Journal of

Risk and Financial

Management

Article
Forecasting Commodity Prices Using the Term Structure
Yasmeen Idilbi-Bayaa and Mahmoud Qadan *

School of Business, Faculty of Social Sciences, University of Haifa, Haifa 3498838, Israel;
[email protected]
* Correspondence: [email protected]

Abstract: The aim of this study is to test the ability of the yield curve on US government bonds
to forecast the future evolution in the prices of commodities often used in as raw materials. We
consider the monthly prices of nine commodities for more than 30 years. Our findings, confirmed by
several parametric and non-parametric tests, are robust and indicate that the ability to forecast future
performance changes over time. Specifically, between 1986 and the early 2000s the yield curve was
quite successful in forecasting monthly changes in commodity prices, but that success diminished in
the period following. One possible explanation for this outcome is the increased flow of capital into
the commodity market resulting in stronger correlations with the equity markets and a breakdown
of the obvious relationship between commodities and business cycle. Our findings are important for
asset pricing, commodity traders and policy makers.

Keywords: forecasting; commodity market; metals; term structure; yield spread

JEL Classification: C53; E3; E4; Q02







Citation: Idilbi-Bayaa, Yasmeen, and

1. Introduction
Mahmoud Qadan. 2021. Forecasting
Commodity Prices Using the Term The literature regards the term structure curve, which plots the yield of government
Structure. Journal of Risk and Financial bonds against their maturity, as an indicator with valuable information about the current
Management 14: 585. https:// and future states of the economy (e.g., Harvey 1989; Abdymomunov 2013; Gogas et al. 2015;
doi.org/10.3390/jrfm14120585 McMillan 2021b). The U.S. Federal Reserve, among other policymakers and institutional
market participants, has always looked at the difference between the yields on long- and
Academic Editor: James W. Kolari short-term sovereign bonds as an indication of where the economy is heading. Thus, the
forecasting ability of the yield spread has become something of a stylized fact among
Received: 5 November 2021 macroeconomists.
Accepted: 30 November 2021 Many studies, detailed in the literature review section, have demonstrated the ability
Published: 4 December 2021 of yield spreads to predict future economic situations effectively. They have established
that the spreads contain a great deal of information about future economic activity and are
Publisher’s Note: MDPI stays neutral accurate predictors of economic growth. However, there have been fewer comprehensive
with regard to jurisdictional claims in
attempts to understand the dynamic relationships between the evolution in the prices
published maps and institutional affil-
of commodities and the shape of the yield curve. This question has become particularly
iations.
relevant in the wake of the unconventional monetary policy used in the last two decades,
which has not been employed since the Great Depression during 1930s. Thus, our goal is
to fill this gap by exploring the ability of the term structure to forecast the future evolution
in the prices of commodities. To accomplish this task, we use 30 years of data about nine
Copyright: © 2021 by the authors. commodities often used as raw materials: coal, gold, silver, oil, platinum, palladium, zinc,
Licensee MDPI, Basel, Switzerland. ethanol, and natural gas. We also use various proxies for the yield spread combined from
This article is an open access article 30-year, 10-year, 2-year, 1-year and 3-month interest rates.
distributed under the terms and
Our findings indicate that yield spreads are generally positively correlated with future
conditions of the Creative Commons
changes in the price of commodities. Our results are robust to controlling for real economic
Attribution (CC BY) license (https://
and financial variables. Adding any or all of these potential alternative explanatory
creativecommons.org/licenses/by/
variables only marginally affects the coefficients or their statistical significance. In our
4.0/).

J. Risk Financial Manag. 2021, 14, 585. https://doi.org/10.3390/jrfm14120585 https://www.mdpi.com/journal/jrfm

J. Risk Financial Manag. 2021, 14, 585 2 of 39

analysis, we also utilize the Engle’s (2002) dynamic conditional correlation procedure
(DCC). The overall picture confirms the existence of time-varying correlation between the
yield spread and future price movements in commodities. To determine why this result
emerged, we consider sub-samples that are determined endogenously using the Bai and
Perron (2003) tests. These structural break tests confirm that the positive correlation is
economically and statistically significant mainly for the period prior to 2004. On the other
hand, in the period following (2004–2020) it seems that the yield spread has been unable to
predict changes in commodities in any significant way.
The weakening correlation between the variables of interest may be related to the
massive capital inflows from individual and institutional investors into the commodity
market in the early 2000s (e.g., Tang and Xiong 2012). Many studies justified these inflows
to commodities by their relatively low correlation with financial markets and, accordingly,
the potential diversification benefits (e.g., Gorton and Rouwenhorst 2006; Daskalaki et al.
2017). However, the considerable inclusion of commodities in investors’ portfolios resulted
in the financialization of commodities, yielding a strong correlation between commodity
prices and equity markets (e.g., Hu et al. 2020) and a breakdown of the obvious relationship
between commodities and cyclical phases of the economy.
Lastly, between 1986 and December 2020, we found eight periods during which the
yield spread was negative or equal to zero. The non-parametric tests conducted to track
the future evolution in commodity prices following flat or downward-sloped yield curves
indicate that such situations can be a successful timing to embark on long positions in
several commodities for investors planning to hold for a relatively long period of time.
Recently, with the outbreak of the Coronavirus late in February 2020, the U.S. 1-year yield
was 1.43%, and the 10-year was about flat (1.46%). Tracking the commodity prices in the
few months following indicate significant shrink in prices. However, the prices recovered
sharply after 2–4 quarters. This recent case, among the others observed, confirms that
specific shapes of the yield curve may generate abnormal returns for investors (see Table A1
in the online Appendix A).
The mechanism underlying our conjecture here accords with the empirical evidence
confirming a strong relationship between the business cycle and commodity prices (Labys
et al. 1999; Chevallier et al. 2014). In addition, the literature has established that financial
markets, including the yield curve, move more quickly than real markets (Saar and Yagil
2015). Accordingly, one should observe a causal relationship from this macro-financial
predictor to the commodity market.
The paper contributes to the existing literature in several ways. First, our study sheds
light on the link between yield spreads and long-term prices in the commodity market.
While previous studies have focused on interest rates in level rather than the difference
between long and short-term sovereign bond yields (e.g., Dai and Kang 2021), little is
known about the information content of the term structure for commodities. Second, we
add to the literature documenting the time-varying relationship between real economic
and financial variables by examining the structural breaks in the long-term correlation
between yield spreads and the commodity market (e.g., Chinn and Kucko 2015). Third,
our findings are especially useful for policy makers and central banks because long-term
predictions about commodity prices are essential in targeting inflation and promoting
overall economic stability (e.g., Garner 1989; Orlowski 2017; Fasanya and Awodimila 2020).
Finally, modeling and forecasting future innovations in commodity prices are important for
both market participants and scholars. Predictions in this area play a vital role in portfolio
optimization and risk management. Indeed, investors are attracted to commodities due
to their inflation-hedging properties (e.g., Beckmann and Czudaj 2013; Bampinas and
Panagiotidis 2015; Levine et al. 2018; Umar et al. 2019), and their possible contribution to
diversifying risks (e.g., Gagnon et al. 2020).
The remainder of this study proceeds as follows. Section 2 reviews the literature.
Section 3 describes the data and the construction of the key variables. Section 4 presents the
J. Risk Financial Manag. 2021, 14, 585 3 of 39

methodology. Section 5 details the empirical findings and discusses the results. Section 6
checks the robustness of the findings, and Section 7 concludes.

2. Literature Review
The macroeconomic literature has established that future real economic growth is pos-
itively correlated with lagged interest spreads (e.g., Stock and Watson 1989; Estrella and
Hardouvelis 1991; Plosser and Rouwenhorst 1994; McMillan 2021a). In parallel, the link
between interest rates and commodities has been also investigated and can be classified into
two categories. The first addresses the effect of the interest rate level on commodity prices.
This line of literature has established that commodity prices increase significantly in response
to reductions in real interest rates (Akram 2009; Arango et al. 2011). The second category
explores the effect of shocks in the commodity market (mainly oil prices) on long-term inter-
est rates (e.g., Ioannidis and Ka 2018). Recent studies use the Granger (1969) causality test
and provide evidence that not only do interest rates drive commodity prices, but also that
commodity prices drive income and interest rates (e.g., Harvey et al. 2017). Despite these
extensive efforts, the examination of the ability of yield spreads to predict future innovations
in commodity prices has attracted relatively less attention in the literature.
The literature points to several reasons why yield spreads forecast future real eco-
nomic activity. One reason relates to the expectation theory, according to which when the
yield curve flattens, market participants expect short-term interest rates to fall due to a re-
cession. This expectation translates into a drop in long-term interest rates, as deteriorating
market conditions during recessions might explain the decline in short-term rates. Indeed,
economic depressions are often associated with job loss, increased uncertainty, business
failures, and credit line contractions. Consequently, if people anticipate a slowdown in
economic activity, there will likely be a drop in the demand for credit, which in turn leads
to a decline in long-term interest rates. On the other hand, if market participants anticipate
an upturn in the economy, future short-term interest rates will be expected to rise, leading
to a steepening of the yield curve. Thus, while falling yield spreads preceding recessions
are caused by both aforementioned factors, the decline in expectations about short-term
rates is the more important one (Hamilton and Kim 2002).
The second explanation is related to the countercyclical monetary policy according
to which economic expansion is accompanied by inflation. To control inflation, central
banks follow a countercyclical monetary policy by raising short-term interest rates. Tight
monetary policies are used to stabilize output growth and cause the yield spread to drop.
This measure is aimed at reducing the anticipation of inflation to levels below the current
inflation rate. Consequently, short-term interest rates rise more than long-term interest
rates do, and the yield curve flattens. As real interest rates remain high, spending decreases,
causing an economic slowdown. Estrella (2005) provided a theoretical model wherein the
yield spread explains both output and inflation. The author showed that the predictive
ability of the yield spread depends upon the reaction of the given monetary policy. By the
same logic, in a recessionary economy, central banks will reduce short-term interest rates
as part of a countercyclical monetary policy. Thus, a lower yield spread or a flat yield curve
is a harbinger of economic downturn.
The third explanation of why the yield curve slope is a leading indicator of economic
output is referred to as the inter-temporal consumption model. As per Harvey (1989),
during expansionary periods people have a stable level of consumption, whereas during
recessions, when income is falling, they tend to reduce their consumption. Hence, if people
anticipate a decline in economic activity, they have an incentive to save in the current
period by selling short-term assets and buying bonds, which will ensure a stable income
during the low-income period. As a result, long-term bond prices rise, which in turn
reduces their yields, and short-term bonds trade at increased rates.
Finally, there are various empirical works dealing with the relationship between the factors
affecting macroeconomic fundamentals and commodity prices. Variables such as an increase in
economic activity (e.g., Duarte et al. 2021), economic uncertainty (Qadan and Nama 2018), the
J. Risk Financial Manag. 2021, 14, 585 4 of 39

exchange rate of the dollar (e.g., Churchill et al. 2019) and the market index (Kagraoka 2016) are
capable of affecting commodity prices. Considering the ability of the spread in bond yields to
anticipate future economic activity, it is very important to have some understanding of its role
in providing information about the future prices of commodities.

3. Data
Our sample consists of monthly data on nine commodities—oil, silver, gold, platinum,
palladium, zinc, ethanol, coal, and natural gas—obtained based on the availability of
the data. These commodities are used in many industries as raw materials. Data about
silver and gold come from the Chicago Mercantile Exchange (CME). Data about platinum
palladium and natural gas come from the New York Mercantile Exchange (NYMEX). The
data on zinc and copper come from the London Bullion Market Association (LBMA).
Information about coal comes from the International Exchange (ICE), whereas the data on
WTI oil are taken from the Federal Reserve Bank of St. Louis.
The largest sample period used is that for oil, gold and silver, and ranges from January
1986 to December 2020, while the smallest sample is that for coal and ranges from January
2009 to December 2020. Our starting point for each commodity is simply due to the
availability of information about their prices. We use the International Monetary Fund’s
International Financial Statistics database for the rates for the 3-month, 1-year, 2-year, 10-
year and 30-year Treasury bills. Table 1 reports the descriptive statistics of the key variables
used in this study and outlines the sample period. Panel A reports the descriptive statistics
of the six proxies used to capture the yield spread, while Panel B reports the descriptive
statistics of the commodities employed here.
Table 1. Descriptive Statistics. Panel A—yield rates. Panel B—commodities.

Panel A
Y3M Y1 Y2 Y10 Y30
Mean 3.173 3.455 3.761 4.845 5.343
Median 3.055 3.390 3.920 4.680 5.155
Maximum 9.140 9.570 9.680 9.520 9.610
Minimum 0.010 0.100 0.130 0.620 1.270
Std. Dev. 2.557 2.623 2.653 2.280 2.052
Skewness 0.258 0.237 0.214 0.199 0.162
Kurtosis 1.845 1.824 1.793 1.981 2.046
J-Bera 28.001 28.118 28.716 20.950 17.767
#Obs. 420 420 420 420 420
1986:01 1986:01 1986:01 1986:01 1986:01
Sample
to to to to to
Period
2020:12 2020:12 2020:12 2020:12 2020:12
Panel B
COAL ETHNL GOLD NGAZ OIL PLDM PLTNM SLVR ZINC
Mean 83.85 1.89 730.10 3.75 44.15 459.31 822.34 11.20 1879.99
Median 82.65 1.77 425.55 2.92 31.90 309.75 680.50 6.72 1891.75
Maximum 130.90 3.62 1970.80 13.92 140.97 2508.80 2180.70 48.58 4474.00
Minimum 49.95 0.82 255.00 1.17 11.13 76.35 336.40 3.56 746.75
Std. Dev. 21.12 0.50 480.20 2.24 28.99 440.48 440.90 8.26 807.30
Skewness 0.280 0.672 0.792 1.721 0.879 2.111 0.813 1.418 0.528
Kurtosis 2.142 3.121 2.151 6.416 2.671 8.299 2.611 4.823 2.765
J-Bera 6.30 14.20 56.52 360.54 55.93 797.60 48.57 198.86 13.68
#Obs. 144 187 420 368 420 417 417 420 281
2009:01 2005:06 1986:01 1990:05 1986:01 1986:04 1986:04 1986:01 1997:08
Sample
to to to to to to to to to
Period
2020:12 2020:12 2020:12 2020:12 2020:12 2020:12 2020:12 2020:12 2020:12
Notes: Panel A of the table reports the descriptive statistics of the Treasury yield rates, whereas Panel B reports those of the commodity
prices. Y3M , Y1 , Y2 , Y10 and Y30 , are US treasury yields on 3-month, 1-year, 2-year, 10-year and 30-year bonds, respectively—all denominated
in annual terms.
J. Risk Financial Manag. 2021, 14, 585 5 of 39

4. Method
The empirical economic literature defines the yield spread as the difference between
the yield rates on long-term and short-term government bonds. In fact, there is no precise
theory that defines how the yield spread should be calculated, and the choice of creating
a proxy for the yield spread is somewhat arbitrary. Indeed, the literature provides many
proxies for the yield spread including the difference between the yields on 10-year bonds
and 3-month bonds (Estrella and Hardouvelis 1991), the difference between 10-year and
1-year interest rates (Stock and Watson 1989) and the difference between yields on 30-year
and 3-month bonds (Duffee 1998). Given the mixed definitions of the yield spread, we
utilize as broad a spectrum of bonds as possible, specifically, the differences in the yields on
10-year and 3-month Treasury bonds, 10-year and 1-year bonds, 10-year and 2-year bonds,
30-year and 3-month bonds, 30-year and 1-year bonds and 30-year and 2-year bonds.
We formulate the following model to trace the effect of the current yield spread (at
time t) on the cumulative rate of change in the “h” subsequent months or quarters.

Rt+h = βh0 + βh1 YLong,t − YShort,t + B0 Xt + vt+h

(1)

where  
12
Rt+h = × 100 × (ln( Pt+h ) − ln( Pt )) (2)
h
Rt+h is the rate of change in the price of the commodity in annual terms. If, for example,
h = 1, then Rt+1 captures the cumulative return of one period (say, a quarter) ahead. If
h = 4, then Rt+4 captures the cumulative returns for the coming twelve months (four
quarters). The difference between the
yield rates on long-term and short-term government
bonds is given by YLong,t − YShort,t and vt+h is the forecast error. Given the possibility
that the forecast error might be correlated, we use Newey and West’s (1987) corrected
covariance estimator. The estimated coefficients guarantee consistency in the presence
of both heteroscedasticity and autocorrelation (HAC) of unknown form. Xt denotes a
matrix of additional explanatory variables. In line with the literature, we use the U.S. dollar
exchange rate (Churchill et al. 2019), the S&P500 (Kagraoka 2016), the industrial production
index (Duarte et al. 2021) and the economic policy uncertainty index (Huang et al. 2021).
To depict the dynamic correlation between the current yield spread and the future
price direction of commodities, we use the established multivariate concept of dynamic
conditional correlation generalized autoregressive conditional heteroscedasticity (DCC
GARCH). Engle (2002) developed this state-of-the-art method, which has been used exten-
sively to quantify dynamic relationships over time. In the following, we provide a very
basic description of this methodology.1
The dynamic conditional correlation estimator is an extension of the constant condi-
tional correlation model suggested by Bollerslev (1990). According to Bollerslev’s proce-
p
dure, the correlation matrix ρ is constant. That is, Ht = Dt ρDt , where Dt = diag hi,t
and hi,t represents the i-th univariate (G)ARCH(p, q) process. In other words,

h1t 0 0 ··· 0
 
 0 h2t 0 ··· 0 
..
 
 
Dt = 
 0 0 h3t . 
 (3)
 . .. ..
 ..

. . 0 
0 0 ··· hnt

According to Engle (2002), ρ is allowed to vary in time t. Thus,

Ht ≡ Dt ρt Dt (4)

The correlation matrix is then given by:

J. Risk Financial Manag. 2021, 14, 585 6 of 39

1 q12,t q13,t · · · q1n,t

 
 q21,t
 1 q23,t · · · q2n,t 

ρt =  q31,t
 q32,t 1 · · · q3n,t 
 (5)
 . .. .. .. ..
 ..

. . . . 
qn1,t qn2,t qn3,t ··· 1
The correlation matrix

is a positive definite one because of the positive nature of Ht .
Given that Qt = qij,t , then:

Qt = (1 − α − β) Q + αηt−1 η0t−1 + βQt−1 (6)

p
where ηt = ε i,t / hi,t are the standardized residuals from the (G)ARCH model, Q is a
n × n matrix and represents the unconditional variance matrix of the standardized error
terms ηt and computed as Q = E[ηt × ηt0 ]. α and β are non-negative scalars and satisfy the
mean-revering assumption (i.e., α + β < 1). Qt is a positive definite matrix that determines
the structure of dynamics and Q∗− t
1
normalizes the elements in Qt ;

√1
 
q11t 0 0 ··· 0
0 √1 0 ··· 0
 
q22t
 
 
Q∗− 1
=
 0 0 √1 ··· 0 
(7)
t q33t 
.. .. .. ..
 
.
 
 . . . 
√
0 0 0 ··· qnnt

In order to estimate the parameters of Ht specifically Φ = (α, β), the following

log-likelihood function is maximized:

T  
L(Φ) = −0.5 ∑ n log(2π ) + log(| Ht |) + y0t Ht−1 yt (8)
i =1
5. Empirical Findings
Table 2 reports the estimation results of the reduced form of Equation (1). That
is, future commodity returns (Rt
+h ) are regressed against the current yield spread only.
Rt+h = βh0 + βh1 YLong,t − YShort,t + vth . In this table, we utilize the difference between
10-year and 3-month bond yields (Y10 -Y3M ) as a proxy for the yield spread. We also use
four forecasting horizons (h; h = 1, 2, 3 and 4) where h = 1 indicates forecasting of one
quarter ahead and h = 4 indicates forecasting four quarters ahead. Panel A of the table
reports the estimation results with respect to the entire sample, Panel B covers 1986 to 2003,
and Panel C covers 2004 to 2020.

Table 2. Estimation results of Equation (1) with the Y10 -Y3M indicator. Panel A: entire sample. Panel B: sample period
1986–2003. Panel C: sample period 2004–2020.

Panel A
Forecast
Oil (1986:01–2020:12) Silver (1986:01–2020:12) Gold (1986:01–2020:12)
Horizon
h C (Y10 -Y3M ) R2 N C (Y10 -Y3M ) R2 N C (Y10 -Y3M ) R2 N
10.61 −3.78 5.33 −0.37 4.89 0.08
1 0.003 139 0.000 139 0.000 138
(0.42) (0.56) (0.48) (0.92) (0.24) (0.97)
9.03 −3.08 3.97 0.39 4.74 0.15
2 0.004 138 0.000 138 0.000 138
(0.29) (0.46) (0.45) (0.88) (0.10) (0.92)
7.41 −2.30 1.85 1.40 4.56 * 0.11
3 0.004 137 0.003 137 0.000 137
(0.24) (0.46) (0.67) (0.50) (0.06) (0.93)
4.29 −0.91 −0.23 2.40 4.47 ** 0.06
4 0.001 136 0.014 136 0.000 136
(0.41) (0.72) (0.95) (0.17) (0.04) (0.95)
J. Risk Financial Manag. 2021, 14, 585 7 of 39

Table 2. Cont.

Panel A
Forecast
Platinum (1986:04–2020:12) Palladium (1986:04–2020:12) Zinc (1997:08–2020:12)
Horizon
h C (Y10 -Y3M ) R2 N C (Y10 -Y3M ) R2 N C (Y10 -Y3M ) R2 N
5.23 −1.48 19.38 * −6.14 3.18 −0.02
1 0.001 139 0.011 139 0.000 93
(0.45) (0.66) (0.06) (0.22) (0.75) (0.99)
3.68 −0.72 15.63 ** −3.95 −1.03 2.62
2 0.000 138 0.009 138 0.005 92
(0.44) (0.76) (0.04) (0.28) (0.89) (0.49)
1.27 0.45 10.53 * −1.16 −4.31 4.53
3 0.0005 136 0.001 137 0.021 91
(0.73) (0.80) (0.09) (0.70) (0.51) (0.17)
−0.18 1.17 8.18 0.24 −6.91 6.04 **
4 0.004 136 0.000 136 0.05 90
(0.95) (0.43) (0.14) (0.93) (0.23) (0.04)
Forecast
Ethanol (2005:06–2020:12) Coal (2009:01–2020:12) Natural Gas (1990:05–2020:12)
Horizon
h C (Y10 -Y3M ) R2 N C (Y10 -Y3M ) R2 N C (Y10 -Y3M ) R2 N
3.52 −2.70 −7.04 4.08 4.24 −1.36
1 0.012 62 0.006 48 0.0003 122
(0.86) (0.79) (0.66) (0.59) (0.79) (0.86)
1.49 −2.32 −10.92 5.85 2.11 −0.40
2 0.002 61 0.0.23 47 0.000 121
(0.90) (0.70) (0.37) (0.31) (0.84) (0.94)
−1.69 −0.64 −15.40 7.67 −1.00 1.02
3 0.004 60 0.06 46 0.001 120
(0.84) (0.88) (0.12) (0.10) (0.91) (0.80)
−5.01 0.61 −14.55 7.14 * −3.51 2.39
4 0.001 59 0.06 45 0.005 119
(0.43) (0.85) (0.11) (0.09) (0.61) (0.47)
Panel B
Forecast
Oil (1986:01–2003:12) Silver (1986:01–2003:12) Gold (1986:01–2003:12)
Horizon
h C (Y10 -Y3M ) R2 N C (Y10 -Y3M ) R2 N C (Y10 -Y3M ) R2 N
11.73 −3.38 −6.16 4.15 −4.99 3.60
1 0.003 71 0.019 71 0.025 70
(0.42) (0.63) (0.40) (0.25) (0.37) (0.18)
10.25 −2.58 −8.19 5.85 ** −4.51 3.36 **
2 0.004 71 0.07 71 0.06 71
(0.29) (0.58) (0.13) (0.03) (0.18) (0.04)
10.62 −2.91 −7.71 * 5.28 *** −3.99 2.74 **
3 0.01 71 0.09 71 0.06 71
(0.13) (0.39) (0.07) (0.01) (0.13) (0.03)
9.55 −2.26 −7.57 ** 5.20 *** −3.54 2.40 **
4 0.009 71 0.114 71 0.06 71
(0.12) (0.44) (0.04) (0.004) (0.13) (0.04)
Forecast
Platinum (1986:04–2003:12) Palladium (1986:04–2003:12) Zinc (1997:08–2003:12)
Horizon
h C (Y10 -Y3M ) R2 N C (Y10 -Y3M ) R2 N C (Y10 -Y3M ) R2 N
−1.01 2.90 11.53 −4.63 −17.87 9.02
1 0.009 71 0.007 71 0.10 25
(0.89) (0.44) (0.40) (0.49) (0.11) (0.12)
−3.28 4.35 * 4.58 −0.001 −15.85 * 9.50 **
2 0.046 71 0.000 71 0.18 25
(0.51) (0.07) (0.66) (0.99) (0.06) (0.03)
−4.65 4.74 ** 0.13 2.20 −14.08 ** 8.39 **
3 0.085 70 0.004 71 0.20 25
(0.23) (0.014) (0.99) (0.60) (0.04) (0.02)
−4.98 4.91 *** −1.30 2.99 −13.17 ** 8.38 ***
4 0.112 71 0.009 71 0.26 25
(0.15) (0.004) (0.87) (0.43) (0.03) (0.01)
J. Risk Financial Manag. 2021, 14, 585 8 of 39

Table 2. Cont.

Panel B
Forecast
Natural Gas (1990:05–2003:12)
Horizon
h C (Y10 -Y3M ) R2 N
8.41 1.11
1 0.0002 54
(0.76) (0.93)
−0.57 5.51
2 0.009 54
(0.97) (0.50)
−5.24 7.99
3 0.03 54
(0.69) (0.19)
−7.17 9.42 *
4 0.07 54
(0.48) (0.05)
Panel C
Forecast
Oil (2004:01–2020:12) Silver (2004:01–2020:12) Gold (2004:01–2020:12)
Horizon
h C (Y10 -Y3M ) R2 N C (Y10 -Y3M ) R2 N C (Y10 -Y3M ) R2 N
9.51 −4.22 17.23 −5.16 14.40 ** −3.29
1 0.002 68 0.009 68 0.019 68
(0.67) (0.70) (0.20) (0.44) (0.02) (0.27)
7.77 −3.65 16.94 * −5.45 14.54 *** −3.29
2 0.004 67 0.022 67 0.033 67
(0.59) (0.60) (0.07) (0.23) (0.002) (0.14)
3.90 −1.62 12.35 −2.87 13.92 *** −2.81
3 0.001 66 0.009 66 0.03 66
(0.72) (0.76) (0.11) (0.44) (0.00) (0.14)
−1.66 0.66 8.09 −0.78 13.53 *** −2.64
4 0.0004 65 0.001 65 0.04 65
(0.85) (0.88) (0.20) (0.80) (0.00) (0.12)
Forecast
Platinum (2004:01–2020:12) Palladium (2004:01–2020:12) Zinc (2004:01–2020:12)
Horizon
h C (Y10 -Y3M ) R2 N C (Y10 -Y3M ) R2 N C (Y10 -Y3M ) R2 N
11.84 −6.18 27.41 * −7.64 13.99 −4.68
1 0.017 68 0.016 68 0.008 68
(0.31) (0.29) (0.07) (0.31) (0.29) (0.48)
11.20 −6.21 27.36 ** −8.18 6.80 −0.91
2 0.04 67 0.036 67 0.0005 67
(0.17) (0.12) (0.013) (0.13) (0.52) (0.86)
7.84 −4.25 21.92 ** −4.88 0.98 2.38
3 0.029 66 0.02 66 0.005 66
(0.21) (0.17) (0.02) (0.26) (2.38) (0.59)
5.27 −2.97 18.91 ** −2.94 −3.42 4.67
4 0.023 65 0.01 65 0.022 65
(0.30) (0.23) (0.02) (0.43) (0.67) (0.24)
Forecast
Natural Gas (2004:01–2020:12)
Horizon
h C (Y10 -Y3M ) R2 N
1.95 −4.07
1 0.003 68
(0.92) (0.67)
5.14 −5.95
2 0.012 67
(0.71) (0.38)
3.37 −5.48
3 0.016 66
(0.76) (0.31)
0.49 −4.21
4 0.016 65
(0.96) (0.32)
Notes: The forecast horizon (h) is in quarters. Y10 -Y3M denotes the yield spread calculated as the difference between the yield rates on
10-year and 3-month government bonds. The table reports the estimation results of Equation (1) with the Newey and West (1987) procedure.
The sample period appears separately for each commodity. Figures in parentheses denote estimated standard errors. ***, ** and * denote
statistical significance at the 1%, 5%, and 10% levels, respectively.
J. Risk Financial Manag. 2021, 14, 585 9 of 39

The picture that emerges indicates an insignificant positive tendency of the yield
spread to forecast future changes in the commodities used. Panel B, however, presents a
different picture. Except for oil and palladium, we find that, when considering 1986–2003,
the yield spread has a positive effect on the future prices of the rest of the commodities.
The ability of the current yield spread to predict future innovations in commodity prices is
manifested in both the statically significant positive coefficients and the relatively high R2
(for example, R2 values are 6%, 7%, 11.2%, 11.4% and 26% for gold, natural gas, platinum,
silver and zinc, respectively.). A steeper yield curve is always viewed as an indication
that the growth in future output is about to rise. Thus, the positive correlation detected
indicates that an increase in the slope at time t will have a positive impact on the future
prices of commodities.
While Table 2 regresses the commodity returns against the yield spread only, in Table 3
we present the full estimation of Equation (1) after controlling for additional explanatory
variables. The sample period considered in Table 3 is for 1986–2003. The results for the
entire sample and the period after 2004 appear in Table A2 (in the online Appendix A). The
overall picture is maintained as evident by the significant positive coefficients of the yield
spreads even after controlling for real and financial economic variables in the period prior
to 2004. The results hold true for all commodities except for oil and palladium.

Table 3. Estimation results of Equation (1) for the period 1986–2003.

Forecast
Horizon Oil (1986:01–2003:12) Silver (1986:01–2003:12)

(Y10 - ∆SP ∆EX ∆IP ∆EPU (Y10 - ∆SP ∆EX ∆IP ∆EPU
h C
Y3M ) R2 N C
Y3M ) R2 N

11.80 −4.97 0.48 −2.20 11.89 0.03 −0.06 71 −8.22 4.93 0.58 −10.74 −1.92 0.00 −0.04 71
1 (0.76) (−0.62) (0.06) (−0.11) (0.67) (0.26) (−1.06) (1.23) (0.15) (−1.06) (−0.22) (−0.07)
10.61 −3.25 −9.41c −5.24 10.78 −0.02 0.00 71 −9.31 6.50b 0.58 −5.78 −3.10 −0.01 0.01 71
2 (1.06) (−0.63) (−1.82) (−0.4) (0.95) (−0.19) (−1.64) (2.22) (0.2) (−0.78) (−0.48) (−0.27)
10.22 −2.73 −5.10 −4.14 3.21 0.00 −0.03 71 −9.37b 6.04a 0.53 −9.09 −2.79 −0.02 0.06 71
3 (1.39) (−0.72) (−1.35) (−0.43) (0.38) (−0.03) (−2.11) (2.64) (0.23) (−1.57) (−0.55) (−0.4)
8.01 −1.83 −2.85 −7.46 2.58 0.02 −0.03 71 −8.92b 5.92a −0.02 −7.43 −2.73 −0.01 0.09 71
4 (1.25) (−0.56) (−0.86) (−0.9) (0.35) (0.44) (−2.36) (3.05) (−0.01) (−1.51) (−0.63) (−0.46)
Forecast
Horizon Gold (1986:01–2003:12) Platinum (1986:04–2003:12)

(Y10 - ∆SP ∆EX ∆IP ∆EPU (Y10 - ∆SP ∆EX ∆IP ∆EPU
h C
Y3M ) R2 N C
Y3M ) R2 N

−7.28 3.49 4.47 −10.20 3.19 0.00 0.01 70 −2.64 2.25 −1.79 −22.72b 7.66 −0.10 0.06 71
1 (−1.23) (1.17) (1.52) (−1.39) (0.5) (0.1) (−0.34) (0.57) (−0.45) (−2.26) (0.87) (−1.57)
−4.01 4.01b −1.50 3.21 −5.20 −0.02 0.04 70 −5.10 4.63c −1.73 −14.33b 2.42 −0.03 0.07 71
2 (−1.14) (2.23) (−0.83) (0.7) (−1.3) (−0.72) (−1.01) (1.78) (−0.66) (−2.17) (0.42) (−0.75)
−3.74 3.04b −0.49 0.67 −2.80 −0.02 0.02 70 −5.71 4.17b 0.32 −10.25c 5.89 −0.02 0.09 71
3 (−1.35) (2.14) (−0.35) (0.19) (−0.89) (−0.93) (−1.42) (2.02) (0.16) (−1.97) (1.29) (−0.49)
−3.62 2.94b −0.87 −0.42 −3.48 −0.02 0.03 70 −5.19 4.32b −0.21 −5.51 4.59 −0.02 0.09 71
4 (−1.46) (2.31) (−0.68) (−0.13) (−1.24) (−0.84) (−1.44) (2.33) (−0.11) (−1.17) (1.12) (−0.7)
Forecast
Horizon Palladium (1986:04–2003:12) Zinc (1997:08–2003:12)

(Y10 - ∆SP ∆EX ∆IP ∆EPU (Y10 - ∆SP ∆EX ∆IP ∆EPU
h C
Y3M ) R2 N C
Y3M ) R2 N

15.89 −10.17 −14.00b −13.30 38.73 −0.18 0.13 71 −19.18c 5.87 8.85 −27.86 9.80 −0.02 0.07 25
1 (1.2) (−1.49) (−2.05) (−0.77) (2.57) (−1.6) (−1.76) (1.00) (1.59) (−1.15) (0.66) (−0.18)
6.75 −4.82 −3.68 −12.06 32.29a −0.08 0.06 71 −17.44b 7.48 5.42 −26.55 4.01 0.04 0.15 25
2 (0.65) (−0.9) (−0.68) (−0.89) (2.72) (−0.9) (−2.13) (1.7) (1.3) (−1.47) (0.36) (0.38)
3.16 −4.11 3.39 −4.46 37.41a −0.02 0.13 71 −15.76b 7.29c 3.40 −24.21 −2.48 0.10 0.21 25
3 (0.38) (−0.95) (0.78) (−0.41) (3.9) (−0.35) (−2.4) (2.07) (1.02) (−1.67) (−0.28) (1.26)
0.88 −2.11 3.50 −4.59 30.30a −0.01 0.10 71 −15.02b 7.85b 2.03 −12.48 1.41 0.09 0.22 25
4 (0.11) (−0.53) (0.88) (−0.46) (3.46) (−0.23) (−2.59) (2.52) (0.69) (−0.97) (0.18) (1.37)
J. Risk Financial Manag. 2021, 14, 585 10 of 39

Table 3. Cont.

Forecast
Risk Financial Manag. 2021, Natural
J. Horizon 14, 585 gas (1990:05–2003:12) 10 of 40
(Y10 - ∆SP ∆EX ∆IP ∆EPU
h C
Y3M ) R2 N

4.31 −3.20 18.12 −12.32 36.33 0.04 −0.04 54

1 −6.06
(0.15) 4.49
(−0.23) (1.33)4.89(−0.33)
−34.36
(1.14) 21.29
(0.16) −0.03 −0.03 54
2 −(−0.34)
6.06 4.49 4.89(0.57)−34.36 −0.03 (−0.18)
21.29 (1.06) −0.03 54
2 (0.52) (−1.43)
(−0.34) (0.52) (0.57) (−1.43) (1.06) (−0.18)
−−16.59
16.59 9.11
9.11 5.02 5.02−50.56a
−50.56a
20.50 20.500.12 0.120.16 0.16
54 54
33 (− 1.32)
(−1.32) (1.51)
(1.51) (0.83)(−3.02)
(0.83) (1.46)(1.46)
(−3.02) (1.15) (1.15)
−15.38 10.68b 3.90 −32.98b 10.68 0.11 0.13 54
4 (−−15.38
1.51) 10.68b
(2.19) (0.8)3.90(−2.44)
−32.98b
(0.94) 10.68
(1.33) 0.11 0.13 54
4
Notes:(−1.51) (2.19)
The forecast horizon(0.8) (−2.44) Y(0.94)
(h) is in quarters. (1.33)
10 -Y3M denotes the yield spread calculated as the difference between the yield rates on
Notes: and
10-year The3-month
forecastgovernment
horizon (h) is inThe
bonds. quarters. Y10-Ythe
table reports denotes the
3M estimation yield
results spread calculated
of Equation as the and
(1) with the Newey difference between
West (1987) the
procedure.
The
yieldsample
rates period appears
on 10-year andseparately
3-monthfor each commodity.
government bonds. Figures in parentheses
The table reports thedenote estimatedresults
estimation standard
of errors.
Equationa, b (1)
andwith
c denote
the
statistical significance at the 1%, 5%, and 10% levels, respectively.
Newey and West (1987) procedure. The sample period appears separately for each commodity. Figures in parentheses
denote estimated standard errors. a, b and c denote statistical significance at the 1%, 5%, and 10% levels, respectively.
The statistically significant positive coefficients accord with the empirical literature
The statistically
postulating significant
that commodity positive
prices coefficients
(specifically accord
silver, gold,with the empirical
platinum, zinc and literature
natural
postulating
gas) are tightlythatlinked
commodity
with the prices (specifically
business cycle andsilver, gold,ofplatinum,
the state the economyzinc (e.g.,
and natural
Batten
et al.are
gas) 2010; Kucher
tightly andwith
linked McCoskey 2017; cycle
the business Jahanandandthe
Serletis 2019).
state of This economically
the economy (e.g., Batten and
et
statistically
al. 2010; Kuchersignificant relationship
and McCoskey 2017;confirms
Jahan andthat these2019).
Serletis metals areeconomically
This used extensivelyand sta-in
various industries,
tistically making them
significant relationship more exposed
confirms to metals
that these the expected
are usedphases in the in
extensively economic
various
cycle.
industries, making them more exposed to the expected phases in the economic cycle.
In investigating
In investigating whether
whether the the relationship
relationship between
between yield
yield spreads
spreads and
and future
future innova-
innova-
tions in commodity
commodityprices pricesisisstable
stableover
overtime, wewe
time, mademadetwotwo
major findings.
major TheThe
findings. first first
is that
is
the correlation
that between
the correlation commodity
between commodity prices and yield
prices spreads
and yield is notisstable
spreads over over
not stable time.time.
This
finding
This is evident
finding in Engle’s
is evident (2002)(2002)
in Engle’s dynamic conditional
dynamic correlations
conditional depicted
correlations in Figures
depicted 1–3,
in Fig-
which illustrate the dynamic conditional correlation between the yield
ures 1–3, which illustrate the dynamic conditional correlation between the yield spreads spreads and the
commodity prices two, three and four quarters ahead. This finding
and the commodity prices two, three and four quarters ahead. This finding accords with accords with recent
studiesstudies
recent maintaining time-varying
maintaining relationship
time-varying between between
relationship yield spread
yieldand futureand
spread economic
future
output (e.g.,
economic Kuosmanen
output et al. 2019;etChinn
(e.g., Kuosmanen andChinn
al. 2019; Kuckoand 2015).
Kucko 2015).

Silver Gold Oil

0.8 1.2 1.2

0.8 0.8
0.4

0.4 0.4
0.0

0.0 0.0

-0.4
-0.4
-0.4

-0.8 -0.8
-0.8

-1.2
-1.2 -1.2
1990 1995 2000 2005 2010 2015 2020
1990 1995 2000 2005 2010 2015 2020 1990 1995 2000 2005 2010 2015 2020

Platinum Palladium Zinc

1.2 1.2 1.2

0.8
0.8 0.8

0.4
0.4 0.4
0.0

0.0 0.0
-0.4

-0.4
-0.4 -0.8

-0.8 -1.2
-0.8 98 00 02 04 06 08 10 12 14 16 18 20
1990 1995 2000 2005 2010 2015 2020
1990 1995 2000 2005 2010 2015 2020

Figure 1. Cont.
J. Risk Financial Manag. 2021, 14, 585 11 of 40
J. Risk Financial Manag. 2021, 14, 585 11 of 40
J. Risk Financial Manag. 2021, 14, 585 11 of 39

Ethanol
Ethanol Coal
Coal NaturalGas
Natural Gas
1.2 1.2 1.2
1.2 1.2 1.2

0.8 0.8
0.8 0.8 0.8
0.8

0.4 0.4
0.4 0.4
0.4
0.4
0.0 0.0
0.0 0.0
0.0
0.0
-0.4 -0.4
-0.4 -0.4
-0.4
-0.4
-0.8-0.8 -0.8
-0.8

-1.2-1.2 -0.8
-0.8
-1.2
-1.2
05 0506 06 0707 0808 0909 1010 1111 1212 1313 1414 1515 1616 1717 18 90 92
92 94
94 9696 9898 0000 0202 0404 0606 0808 1010 1212 1414 1616 1818 20 20
18 19
19 20
20 09
09 10
10 11
11 12
12 13
13 14
14 15
15 16
16 17
17 18
18 19
19 20
20
90

Figure 1. Dynamic correlation between the yield spread and two quarters ahead.
Figure
Figure1.1.Dynamic
Dynamic correlation
correlation between
between the yield spread
spread and
and two
two quarters
quartersahead.
ahead.

Silver
Silver Gold Oil
Oil
1.21.2 1.2
1.2 1.2
1.2

0.8
0.8 0.8
0.8
0.80.8

0.4
0.4 0.4
0.4
0.40.4

0.0 0.0
0.0
0.00.0 0.0

-0.4
-0.4
-0.4
-0.4
-0.4-0.4

-0.8
-0.8
-0.8
-0.8
-0.8-0.8
-1.2
-1.2
-1.2
-1.2 1990 1995 2000 2005 2010 2015 2020
-1.2-1.2 1990 1995 2000 2005 2010 2015 2020
1990 1995 2000 2005 2010 2015 2020
1990
1990 1995
1995 2000
2000 2005
2005 2010
2010 2015
2015 2020
2020 1990 1995 2000 2005 2010 2015 2020

Platinum
Platinum Palladium
Palladium Zinc
Zinc
1.2 1.2 1.2
1.2 1.2 1.2

0.8 0.8
0.8 0.8 0.8
0.8
0.4 0.4
0.4 0.4
0.4
0.4
0.0 0.0
0.0 0.0

0.0
0.0 -0.4
-0.4 -0.4
-0.4

-0.4 -0.8
-0.8 -0.8
-0.4 -0.8
-1.2
-0.8 -1.2 -1.2 98 00 02 04 06 08 10 12 14 16 18 20
-0.8 -1.2 1990 1995 2000 2005 2010 2015 2020
1990 1995 2000 2005 2010 2015 2020 98 00 02 04 06 08 10 12 14 16 18 20
1990 1995 2000 2005 2010 2015 2020 1990 1995 2000 2005 2010 2015 2020

Ethanol
Ethanol Coal Natural
Coal Natural Gas
Gas
1.2
1.2 1.2
1.2
1.2 1.2
1.0
1.0 0.8
0.8
0.8
0.8 0.8
0.8
0.6 0.4
0.4
0.6 0.4
0.4
0.4
0.4 0.0 0.0
0.2 0.0
0.0
0.2
0.0 -0.4
-0.4
0.0 -0.4
-0.2 -0.4
-0.8
-0.2 -0.8
-0.4 -0.8
-0.8
-0.4 -1.2
-0.6 -1.2
05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 -1.290 92 94 96 98 00 02 04 06 08 10 12 14 16 18 20
-0.6 -1.2 09 10 11 12 13 14 15 16 17 18 19 20 90 92 94 96 98 00 02 04 06 08 10 12 14 16 18 20
05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 09 10 11 12 13 14 15 16 17 18 19 20

Figure 2. Dynamic correlation between the yield spread and three quarters ahead.
Figure 2. Dynamic correlation between the yield spread and three quarters ahead.
Figure 2. Dynamic correlation between the yield spread and three quarters ahead.
J. Risk Financial Manag. 2021, 14, 585 12 of 39
J. Risk Financial Manag. 2021, 14, 585 12 of 40

Silver Gold Oil

1.2 1.2
1.2

0.8 0.8
0.8

0.4
0.4 0.4

0.0
0.0 0.0

-0.4
-0.4 -0.4

-0.8
-0.8 -0.8

-1.2
-1.2 -1.2 1990 1995 2000 2005 2010 2015 2020
1990 1995 2000 2005 2010 2015 2020 1990 1995 2000 2005 2010 2015 2020

Platinum Palladium Zinc

1.2 1.2 1.2

0.8 0.8 0.8

0.4 0.4 0.4

0.0 0.0 0.0

-0.4
-0.4 -0.4

-0.8
-0.8 -0.8

-1.2
-1.2 -1.2
98 00 02 04 06 08 10 12 14 16 18
1990 1995 2000 2005 2010 2015 2020 1990 1995 2000 2005 2010 2015 2020

Ethanol Coal Natural Gas

0.4 1.2 1.2

0.2
0.8 0.8
0.0
0.4 0.4
-0.2

-0.4 0.0 0.0

-0.6
-0.4 -0.4
-0.8

-0.8 -0.8
-1.0

-1.2 -1.2 -1.2

05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 90 92 94 96 98 00 02 04 06 08 10 12 14 16 18
09 10 11 12 13 14 15 16 17 18 19 20

Figure 3. Dynamic correlation between the yield spread and four quarters ahead.
Figure 3. Dynamic correlation between the yield spread and four quarters ahead.
Second, using the Bai and Perron (2003) structural break test we find that the early
2000sSecond, using associated
is the period the Bai andwith
Perron (2003) breaks
structural structural break
in the test we find
relationship that the
between theearly
yield
2000s and
curve is thethe
period
pricesassociated with structural
of commodities. Table 4 breaks
presents in the
the results
relationship
of thisbetween
test andthe yield
indicates
curve
the anddetected
dates the pricesasofstructural
commodities.
breakTable 4 presents
points. the results
In general, of this test
the findings andand
of Bai indicates
Perron
the dates
(2003) test detected as structural
point to the 2003–3004break
as thepoints.
period Inin general,
which therethe was
findings of Bai and
a structural breakPerron
in the
(2003) test point to the 2003–3004 as the period in which there was
relationship between yield spreads and future commodity prices. These findings accord a structural break in
the relationship
with earlier studiesbetween yield
that date thespreads
start of and future commodity
the financialization prices. These
commodities findings
to the ac-
early 2000s
cord with earlier
(Hamilton and Wustudies
2015;that date the start
Henderson et al.of2015).
the financialization
In other words, commodities
this period tomarks
the earlythe
2000sof(Hamilton
start and Wu
the increased 2015; Henderson
exposure of portfolio et managers,
al. 2015). Inindividuals
other words,and thishedge
periodfunds
marksto
the start of the increased exposure of portfolio managers, individuals and hedge funds to
commodities.
commodities.
Table 4. Bai and Perron (2003) multiple break-point test.

h = 2 (Two Quarters Ahead)

Silver Oil Gold PLTNM PLDM Zinc ETHNL Coal NatGas
Break Point #1 2003Q2 = 2001Q2 = = = = = =
Break Point #2 2011Q2 = 2012Q3 = = = = = =
Break Point #3 = = = = = = = = =
J. Risk Financial Manag. 2021, 14, 585 13 of 39

Table 4. Cont.

h = 3 (Three Quarters Ahead)

Silver Oil Gold PLTNM PLDM Zinc ETHNL Coal NatGas
Break Point #1 2003Q2 = 2001Q1 = 1996Q3 = = 2011Q1 =
Break Point #2 2011Q1 = 2012Q2 = = = = 2015Q4 =
Break Point #3 = = = = = = = = =
h = 4 (Four Quarters Ahead)
Silver Oil Gold PLTNM PLDM Zinc ETHNL Coal NatGas
Break Point #1 2003Q1 1995Q4 2001Q2 1998Q4 1996Q2 = = 2011Q1 2000Q3
Break Point #2 2011Q1 = 2012Q2 2010Q4 2001Q2 = = 2015Q4 2006Q2
Break Point #3 = = = = 2008Q4 = = = =
Notes: We tested for any structural break using the Bai and Perron (2003) multiple break-point test. The values listed in the table are those
of the break dates. The vast majority of the commodities point to the 2000s as the structural break points. “=” denotes that no significant
breakpoint was detected between the future return and the current yield spread.

This finding is in line with prior studies documenting the weakening ability of the term
structure to predict future economic activity. Early on, Stock and Watson (2003) and Giacomini
and Rossi (2006) maintained that the yield spread’s ability to forecast economic expansion has
weakened since the 1980s, but its predictive ability remains strong only for recessions. Other
works raise questions regarding the stability of the term spread’s predictive content (e.g.,
Wheelock and Wohar 2009). Evgenidis et al. (2020) confirm the time-varying nature of the
yield spread’s predictive ability, mainly during the 2000s.
One possible factor explaining this break between commodities and the most reliable
indicator of future economic activity is the financialization of commodities. For a long time,
commodities were viewed as a segmented market offering significant diversification benefits in
light of the low—even negative—correlation between their returns and the stock market (e.g.,
Bodie and Rosansky 1980; Demiralay et al. 2019). This characteristic prompted traders, financial
institutions and institutional investors to consider this new asset class as a useful diversifier in
their portfolios. A byproduct of this development is the acceleration in the financialization of
these commodities, which in turn fueled a rapid increase in their co-movements with equity
markets (e.g., Qadan et al. 2019). This evolution may explain the breakdown of the obvious
relationship between commodities and the expected economic evolution.
We also test the extent to which the dynamic correlation of commodity “i” co-moves
with that of commodity “j.” A quick glance at Figures 1–3 shows the apparent co-movements
between some of these commodities. Table 5 presents the simple correlation between the
DCC values. Some of the correlation values are negative and statistically significant. For
example, we detect a negative correlation between the prices of ethanol and gold, gold
and natural gas, natural gas and palladium, oil, and palladium. On the other hand, the
majority of the other cases are associated with statistically significant positive correlations,
particularly for precious metals. For example, the correlation between the DCC values
of gold and silver is 0.73, and that between gold and platinum is 0.534. Overall, this
picture reveals that the conditional slope among commodities is largely connected—a clear
indication of their similar reaction to the current yield spread.

Table 5. Pearson’s correlation between the DCC values of the commodities.

Coal Ethanol Gold Nat.Gas Oil PLDM PLTNM Silver Zinc

Coal 1.00
Ethanol 0.647 *** 1.00
Gold 0.090 −0.230 * 1.00
Nat.Gas 0.124 0.321 ** −0.153 * 1.00
Oil 0.692 *** 0.519 *** 0.167 * 0.627 *** 1.00
PLDM 0.407 *** 0.205 0.166 * −0.222 ** −0.141 * 1.00
J. Risk Financial Manag. 2021, 14, 585 14 of 39

Table 5. Cont.

Coal Ethanol Gold Nat.Gas Oil PLDM PLTNM Silver Zinc

PLTNM 0.120 −0.037 0.534 *** 0.354 *** 0.406 *** 0.327 *** 1.00
Silver 0.249 −0.071 0.730 *** −0.065 0.091 0.487 *** 0.630 *** 1.00
Zinc 0.493 *** 0.277 ** −0.154 0.086 0.196 * 0.278 *** 0.081 0.159 1.00
Notes: ***, ** and * denote statistical significance at the 1%, 5%, and 10% levels, respectively.

6. Robustness Checks
6.1. Additional Proxies for Yield
Previous studies suggested capturing the yield spread using different proxies. To
develop a broader picture regarding the interaction between the yield spread and the
future evolution in commodity prices, we depart from the standard yield spread used in
the literature (Y10 -Y3M ) and test other proxies: the difference between 10-year and 1-year
interest rates (Y10 -Y1 ), between 10-year and 2-year yields (Y10 -Y2 ), between 30-year and
3-months yields (Y30 -Y3M ), between 30-year and 1-year bond yields (Y30 -Y1 ) and between
30-year and 2-year bond yields (Y30 -Y2 ).
Tables 6–10 report the estimation results of the prediction model. Table 6 presents
the estimation results given Y10 -Y1 as the yield spread. In Table 7, Y10 -Y2 proxies for the
yield spread. In Table 8, Y30 -Y3M proxies for the yield spread. Table 9 utilizes Y30 -Y1 as
the yield spread, and Table 10 utilizes Y30 -Y2 to proxy for the yield spread. The overall
picture is maintained as evident by the significant positive coefficients in the period prior to
2004 (Panel B in each table), but the insignificant results in the period that follows (Panels
C). The regression results that include the other explanatory variables reflect very similar
picture. They appear in Tables A3–A7 in the online Appendix A.
A closer glance at the results in Panel B of Table 6 confirms that the yield spread,
defined as Y10 -Y1 , is an efficient predictor of the future prices of silver, gold, platinum
and zinc. The resulting R2 for silver ranges between 0.03 when forecasting one quarter
ahead (h = 1), and 0.19 when forecasting the prices one year ahead (h = 4). We find that
the regression R2 for gold ranges between 0.04 (for h = 1) and 0.15 (for h = 4), for platinum
it ranges between 0.01 (for h = 1) and 0.12 (for h = 4), and finally it ranges between 0.09
(for h = 1) and 0.26 for zinc (h = 4). This picture is essentially replicated in Tables 7–10.
Moreover, Panel B of Table 10 provides strong support for these findings. The resulting R2
for zinc ranges between 0.08 (for h = 1), and 0.34 (for h = 4). By and large, these findings
confirm the premise that metal prices are positively correlated with macroeconomic activity
(e.g., Fama and French 1988).

Table 6. Estimation results of Equation (1) with the Y10 -Y1 indicator. Panel A: entire sample. Panel B: sample period
1986–2003. Panel C: sample period 2004–2020.

Panel A
Forecast
Oil (1986:01–2020:12) Silver (1986:01–2020:12) Gold (1986:01–2020:12)
Horizon
h C (Y10 -Y1 ) R2 N C (Y10 -Y1 ) R2 N C (Y10 -Y1 ) R2 N
10.51 −4.45 3.28 1.01 3.82 0.84
1 0.003 139 0.000 139 0.001 138
(0.25) (0.38) (0.61) (0.79) (0.40) (0.72)
8.59 −3.40 2.60 1.43 3.65 0.94
2 0.004 138 0.002 138 0.003 138
(0.30) (0.47) (0.67) (0.70) (0.41) (0.68)
7.34 −2.71 1.03 2.26 3.41 0.94
3 0.005 137 0.007 137 0.004 137
(0.35) (0.55) (0.86) (0.52) (0.42) (0.67)
4.95 −1.56 −0.65 3.19 3.30 0.89
4 0.002 136 0.021 136 0.004 136
(0.51) (0.72) (0.90) (0.35) (0.42) (0.67)
J. Risk Financial Manag. 2021, 14, 585 15 of 39

Table 6. Cont.

Panel A
Forecast
Platinum (1986:04–2020:12) Palladium (1986:04–2020:12) Zinc (1997:08–2020:12)
Horizon
h C (Y10 -Y1 ) R2 N C (Y10 -Y1 ) R2 N C (Y10 -Y1 ) R2 N
3.98 −0.88 18.34 ** −6.63 3.24 −0.06
1 0.000 139 0.01 139 0.000 93
(0.45) (0.78) (0.04) (0.25) (0.80) (0.99)
3.15 −0.49 15.69 * −4.79 0.19 2.13
2 0.0003 138 0.01 138 0.003 92
(0.54) (0.87) (0.08) (0.39) (0.99) (0.71)
1.71 0.23 12.24 −2.59 −2.26 3.74
3 0.0001 136 0.004 137 0.01 91
(0.73) (0.94) (0.18) (0.63) (0.84) (0.46)
0.49 0.94 10.16 −1.10 −4.33 5.12
4 0.002 136 0.001 136 0.03 90
(0.92) (0.72) (0.27) (0.83) (0.69) (0.27)
Forecast
Ethanol (2005:06–2020:12) Coal (2009:01–2020:12) Natural gas (1990:05–2020:12)
Horizon
h C (Y10 -Y1 ) R2 N C (Y10 -Y1 ) R2 N C (Y10 -Y1 ) R2 N
2.00 −1.97 −4.91 3.21 5.07 −2.16
1 0.001 62 0.004 48 0.001 122
(0.88) (0.80) (0.79) (0.71) (0.71) (0.76)
0.20 −1.70 −8.31 4.90 2.75 −0.90
2 0.001 61 0.02 47 0.0002 121
(0.98) (0.78) (0.61) (0.52) (0.82) (0.89)
−1.68 −0.72 −12.06 6.51 0.96 −0.12
3 0.000 60 0.04 46 0.000 120
(0.84) (0.90) (0.40) (0.34) (0.93) (0.98)
−4.84 0.56 −11.26 5.98 −1.10 1.19
4 0.000 59 0.04 45 0.001 119
(0.46) (0.91) (0.42) (0.37) (0.91) (0.82)
Panel B
Forecast
Oil (1986:01–2003:12) Silver (1986:01–2003:12) Gold (1986:01–2003:12)
Horizon
h C (Y10 -Y1 ) R2 N C (Y10 -Y1 ) R2 N C (Y10 -Y1 ) R2 N
12.42 −4.79 −7.50 6.26 * −5.59 5.00 **
1 0.005 71 0.03 71 0.04 70
(0.31) (0.43) (0.15) (0.07) (0.14) (0.03)
10.64 −3.56 −9.37 * 8.28 ** −5.59 * 5.06 ***
2 0.006 71 0.11 71 0.10 71
(0.33) (0.50) (0.05) (0.03) (0.07) (0.01)
9.74 −3.03 −8.99 ** 7.64 ** −5.41 ** 4.52 ***
3 0.009 71 0.14 71 0.13 71
(0.32) (0.53) (0.03) (0.01) (0.04) (0.004)
8.86 −2.35 −8.93 ** 7.60 *** −5.18 ** 4.26 ***
4 0.007 71 0.19 71 0.15 71
(0.34) (0.62) (0.01) (0.003) (0.04) (0.003)
Forecast
Platinum (1986:04–2003:12) Palladium (1986:04–2003:12) Zinc (1997:08–2003:12)
Horizon
h C (Y10 -Y1 ) R2 N C (Y10 -Y1 ) R2 N C (Y10 -Y1 ) R2 N
−0.64 3.39 14.98 −8.42 −16.75 9.49
1 0.01 71 0.02 71 0.09 25
(0.93) (0.35) (0.29) (0.31) (0.14) (0.12)
−2.72 5.09 8.46 −2.89 −14.68 10.01 *
2 0.05 71 0.004 71 0.17 25
(0.68) (0.12) (0.54) (0.72) (0.15) (0.09)
−3.93 5.45 * 4.74 −0.63 −13.22 8.97 *
3 0.09 70 0.000 71 0.20 25
(0.52) (0.07) (0.73) (0.93) (0.18) (0.09)
−4.48 5.85 * 2.72 0.80 −12.50 9.12 *
4 0.12 71 0.001 71 0.26 25
(0.43) (0.04) (0.84) (0.91) (0.17) (0.06)
J. Risk Financial Manag. 2021, 14, 585 16 of 39

Table 6. Cont.

Panel B
Forecast
Natural Gas (1990:05–2003:12)
Horizon
h C (Y10 -Y1 ) R2 N
10.73 −0.22
1 0.000 54
(0.67) (0.99)
2.97 4.30
2 0.004 54
(0.89) (0.66)
−0.41 6.44
3 0.02 54
(0.98) (0.47)
−1.84 7.84
4 0.04 54
(0.92) (0.30)
Panel C
Forecast
Oil (2004:01–2020:12) Silver (2004:01–2020:12) Gold (2004:01–2020:12)
Horizon
h C (Y10 -Y1 ) R2 N C (Y10 -Y1 ) R2 N C (Y10 -Y1 ) R2 N
8.26 −3.95 15.22 −4.46 13.82 ** −3.34
1 0.002 68 0.006 68 0.02 68
(0.60) (0.64) (0.10) (0.46) (0.01) (0.29)
6.04 −2.99 15.87 * −5.53 14.21 *** −3.52
2 0.002 67 0.019 67 0.03 67
(0.65) (0.69) (0.06) (0.34) (0.01) (0.26)
4.29 −2.10 12.56 −3.41 13.88 *** −3.18
3 0.002 66 0.012 66 0.04 66
(0.73) (0.76) (0.12) (0.55) (0.01) (0.29)
−0.09 −0.28 9.14 −1.59 13.71 *** −3.12
4 0.000 65 0.004 65 0.05 65
(0.99) (0.97) (0.25) (0.78) (0.004) (0.26)
Forecast
Platinum (2004:01–2020:12) Palladium (2004:01–2020:12) Zinc (2004:01–2020:12)
Horizon
h C (Y10 -Y1 ) R2 N C (Y10 -Y1 ) R2 N C (Y10 -Y1 ) R2 N
8.60 −4.83 22.90 ** −5.63 13.37 −4.91
1 0.01 68 0.01 68 0.01 68
(0.24) (0.30) (0.01) (0.44) (0.46) (0.58)
9.07 −5.64 24.29 *** −7.25 7.93 −1.80
2 0.03 67 0.02 67 0.002 67
(0.19) (0.23) (0.004) (0.32) (0.65) (0.82)
7.55 −4.66 21.46 ** −5.25 3.58 0.97
3 0.03 66 0.02 66 0.000 66
(0.25) (0.27) (0.01) (0.43) (0.83) (0.89)
5.67 −3.65 19.62 ** −3.82 0.15 2.97
4 0.03 65 0.01 65 0.01 65
(0.37) (0.33) (0.03) (0.54) (0.99) (0.66)
Forecast
Natural Gas (2004:01–2020:12)
Horizon
h C (Y10 -Y1 ) R2 N
0.56 −3.69
1 0.002 68
(0.97) (0.63)
2.48 −4.99
2 0.01 67
(0.86) (0.49)
1.93 −5.27
3 0.01 66
(0.88) (0.41)
−0.77 −3.97
4 0.01 65
(0.94) (0.47)
Notes: The forecast horizon (h) is in quarters. Y10 -Y1 denotes the yield spread calculated as the difference between the yield rates on
10-year and 1-year government bonds. The table reports the estimation results of Equation (1) with the Newey and West (1987) procedure.
The sample period appears separately for each commodity. Figures in parentheses denote estimated standard errors. ***, ** and * denote
statistical significance at the 1%, 5%, and 10% levels, respectively.
J. Risk Financial Manag. 2021, 14, 585 17 of 39

Table 7. Estimation results of Equation (1) with the Y10 -Y2 indicator. Panel A: entire sample. Panel B: sample period
1986–2003. Panel C: sample period 2004–2020.

Panel A
Forecast
Oil (1986:01–2020:12) Silver (1986:01–2020:12) Gold (1986:01–2020:12)
Horizon
h C (Y10 -Y2 ) R2 N C (Y10 -Y2 ) R2 N C (Y10 -Y2 ) R2 N
11.27 −6.39 2.30 2.18 2.96 1.86
1 0.004 139 0.001 139 0.004 138
(0.20) (0.30) (0.70) (0.65) (0.48) (0.50)
8.47 −4.25 1.81 2.56 2.87 1.91
2 0.004 138 0.004 138 0.01 138
(0.27) (0.44) (0.75) (0.58) (0.48) (0.47)
7.27 −3.41 0.69 3.20 2.61 1.92
3 0.005 137 0.01 137 0.01 137
(0.32) (0.52) (0.90) (0.47) (0.51) (0.45)
5.27 −2.29 −0.55 3.99 2.50 1.86
4 0.003 136 0.02 136 0.01 136
(0.45) (0.66) (0.91) (0.35) (0.51) (0.45)
Forecast
Platinum (1986:04–2020:12) Palladium (1986:04–2020:12) Zinc (1997:08–2020:12)
Horizon
h C (Y10 -Y2 ) R2 N C (Y10 -Y2 ) R2 N C (Y10 -Y2 ) R2 N
3.51 −0.71 16.64 * −6.94 3.79 −0.54
1 0.0002 139 0.01 139 0.000 93
(0.50) (0.85) (0.06) (0.33) (0.76) (0.94)
2.90 −0.40 15.04 * −5.55 1.10 1.78
2 0.0001 138 0.01 138 0.002 92
(0.55) (0.91) (0.09) (0.42) (0.93) (0.78)
2.18 −0.13 12.77 −3.80 −0.98 3.41
3 0.000 136 0.01 137 0.01 91
(0.65) (0.97) (0.15) (0.57) (0.93) (0.56)
1.29 0.49 10.88 −2.05 −2.64 4.73
4 0.0005 136 0.003 136 0.02 90
(0.77) (0.88) (0.21) (0.74) (0.80) (0.38)
Forecast
Ethanol (2005:06–2020:12) Coal (2009:01–2020:12) Natural gas (1990:05–2020:12)
Horizon
h C (Y10 -Y2 ) R2 N C (Y10 -Y2 ) R2 N C (Y10 -Y2 ) R2 N
1.64 −1.94 −4.20 3.23 6.92 −4.28
1 0.0004 62 0.003 48 0.002 122
(0.91) (0.84) (0.82) (0.74) (0.59) (0.62)
−0.35 −1.50 −6.71 4.59 5.23 −3.24
2 0.001 61 0.01 47 0.002 121
(0.97) (0.84) (0.69) (0.62) (0.65) (0.67)
−1.33 −1.11 −9.94 6.12 3.88 −2.62
3 0.001 60 0.03 46 0.002 120
(0.88) (0.87) (0.51) (0.47) (0.72) (0.71)
−4.08 0.05 −8.80 5.31 1.61 −0.77
4 0.000 59 0.03 45 0.0003 119
(0.55) (0.99) (0.56) (0.53) (0.87) (0.90)
Panel B
Forecast
Oil (1986:01–2003:12) Silver (1986:01–2003:12) Gold (1986:01–2003:12)
Horizon
h C (Y10 -Y2 ) R2 N C (Y10 -Y2 ) R2 N C (Y10 -Y2 ) R2 N
12.81 −7.33 −6.76 8.23 * −5.11 6.71 **
1 0.01 71 0.04 71 0.05 70
(0.24) (0.31) (0.17) (0.07) (0.15) (0.02)
10.36 −4.83 −8.65 * 11.18 ** −5.24 * 6.93 ***
2 0.01 71 0.13 71 0.13 71
(0.30) (0.46) (0.06) (0.02) (0.07) (0.004)
8.78 −3.34 −8.73 ** 10.73 *** −5.34 ** 6.45 ***
3 0.01 71 0.18 71 0.18 71
(0.34) (0.57) (0.03) (0.01) (0.03) (0.001)
7.32 −1.74 −8.63 *** 10.65 *** −5.25 ** 6.22 ***
4 0.003 71 0.24 71 0.21 71
(0.39) (0.77) (0.01) (0.001) (0.03) (0.00)
J. Risk Financial Manag. 2021, 14, 585 18 of 39

Table 7. Cont.

Panel B
Forecast
Platinum (1986:04–2003:12) Palladium (1986:04–2003:12) Zinc (1997:08–2003:12)
Horizon
h C (Y10 -Y2 ) R2 N C (Y10 -Y2 ) R2 N C (Y10 -Y2 ) R2 N
−0.19 4.41 14.57 −11.71 −14.39 10.23
1 0.01 71 0.02 71 0.08 25
(0.98) (0.37) (0.25) (0.27) (0.22) (0.17)
−1.92 6.49 9.81 −5.60 −13.13 11.82 *
2 0.05 71 0.01 71 0.18 25
(0.77) (0.15) (0.44) (0.58) (0.19) (0.09)
−2.94 6.78 * 6.75 −3.07 −12.20 10.99 *
3 0.09 70 0.004 71 0.22 25
(0.62) (0.09) (0.59) (0.75) (0.18) (0.06)
−3.43 7.32 ** 4.49 −0.74 −11.53 11.25 **
4 0.13 71 0.0003 71 0.29 25
(0.54) (0.05) (0.71) (0.93) (0.16) (0.02)
Forecast
Natural Gas (1990:05–2003:12)
Horizon
h C (Y10 -Y2 ) R2 N
12.09 −1.61
1 0.0002 54
(0.59) (0.91)
7.03 2.16
2 0.001 54
(0.72) (0.85)
3.90 4.94
3 0.01 54
(0.83) (0.64)
1.86 7.49
4 0.02 54
(0.91) (0.40)
Panel C
Forecast
Oil (2004:01–2020:12) Silver (2004:01–2020:12) Gold (2004:01–2020:12)
Horizon
h C (Y10 -Y2 ) R2 N C (Y10 -Y2 ) R2 N C (Y10 -Y2 ) R2 N
9.12 −5.25 14.73 −4.77 13.80 *** −3.85
1 0.002 68 0.005 68 0.02 68
(0.58) (0.61) (0.13) (0.53) (0.01) (0.31)
5.56 −3.09 15.76 * −6.31 14.40 *** −4.22
2 0.002 67 0.02 67 0.03 67
(0.67) (0.71) (0.08) (0.38) (0.01) (0.26)
4.61 −2.68 13.68 −4.82 14.39 *** −4.07
3 0.002 66 0.02 66 0.04 66
(0.71) (0.73) (0.10) (0.49) (0.01) (0.25)
1.36 −1.44 10.82 −3.11 14.36 *** −4.11
4 0.001 65 0.01 65 0.06 65
(0.91) (0.85) (0.17) (0.64) (0.00) (0.22)
Forecast
Platinum (2004:01–2020:12) Palladium (2004:01–2020:12) Zinc (2004:01–2020:12)
Horizon
h C (Y10 -Y2 ) R2 N C (Y10 -Y2 ) R2 N C (Y10 -Y2 ) R2 N
7.53 −4.74 21.92 ** −5.74 15.05 −6.99
1 0.01 68 0.01 68 0.01 68
(0.31) (0.36) (0.02) (0.50) (0.41) (0.51)
8.12 −5.78 23.87 *** −8.05 10.20 −3.84
2 0.02 67 0.02 67 0.01 67
(0.25) (0.26) (0.01) (0.33) (0.57) (0.69)
7.92 −5.67 22.88 *** −7.17 6.35 −1.01
3 0.03 66 0.03 66 0.001 66
(0.23) (0.24) (0.01) (0.36) (0.71) (0.91)
6.56 −4.90 21.70 *** −6.01 3.36 0.99
4 0.04 65 0.03 65 0.001 65
(0.29) (0.25) (0.01) (0.40) (0.84) (0.90)
J. Risk Financial Manag. 2021, 14, 585 19 of 39

Table 7. Cont.

Panel C
Forecast
Natural Gas (2004:01–2020:12)
Horizon
h C (Y10 -Y2 ) R2 N
0.76 −4.42
1 0.002 68
(0.96) (0.63)
2.15 −5.52
2 0.01 67
(0.88) (0.52)
2.29 −6.38
3 0.01 66
(0.86) (0.40)
−0.56 −4.76
4 0.01 65
(0.96) (0.47)
Notes: The forecast horizon (h) is in quarters. Y10 -Y2 denotes the yield spread calculated as the difference between the yield rates on
10-year and 2-year government bonds. The table reports the estimation results of Equation (1) with the Newey and West (1987) procedure.
The sample period appears separately for each commodity. Figures in parentheses denote estimated standard errors. ***, ** and * denote
statistical significance at the 1%, 5%, and 10% levels, respectively.

Table 8. Estimation results of Equation (1) with the Y30 -Y3M indicator. Panel A: entire sample. Panel B: sample period
1986–2003. Panel C: sample period 2004–2020.

Panel A
Forecast
Oil (1986:01–2020:12) Silver (1986:01–2020:12) Gold (1986:01–2020:12)
Horizon
h C (Y30 -Y3M ) R2 N C (Y30 -Y3M ) R2 N C (Y30 -Y3M ) R2 N
10.14 −2.70 3.39 0.60 4.43 0.27
1 0.002 139 0.0003 139 0.0002 138
(0.32) (0.48) (0.63) (0.84) (0.37) (0.88)
8.54 −2.16 2.55 0.95 4.26 0.33
2 0.003 138 0.001 138 0.001 138
(0.36) (0.54) (0.70) (0.73) (0.37) (0.85)
7.38 −1.77 0.96 1.49 3.99 0.34
3 0.004 137 0.01 137 0.001 137
(0.40) (0.60) (0.88) (0.57) (0.38) (0.83)
4.49 −0.80 −0.53 2.00 3.99 0.27
4 0.001 136 0.01 136 0.001 136
(0.59) (0.80) (0.93) (0.42) (0.36) (0.87)
Forecast
Platinum (1986:04–2020:12) Palladium (1986:04–2020:12) Zinc (1997:08–2020:12)
Horizon
h C (Y30 -Y3M ) R2 N C (Y30 -Y3M ) R2 N C (Y30 -Y3M ) R2 N
4.83 −0.96 19.20 * −4.65 1.46 0.76
1 0.001 139 0.01 139 0.0004 93
(0.43) (0.70) (0.07) (0.29) (0.92) (0.87)
3.92 −0.67 16.41 −3.41 −2.14 2.40
2 0.001 138 0.01 138 0.01 92
(0.51) (0.78) (0.11) (0.42) (0.87) (0.57)
1.72 0.15 11.49 −1.33 −5.11 3.66
3 0.000 136 0.002 137 0.02 91
(0.76) (0.95) (0.27) (0.74) (0.68) (0.33)
0.34 0.68 8.62 −0.01 −7.57 4.71
4 0.002 136 0.000 136 0.05 90
(0.95) (0.73) (0.40) (0.99) (0.52) (0.17)
Forecast
Ethanol (2005:06–2020:12) Coal (2009:01–2020:12) Natural gas (1990:05–2020:12)
Horizon
h C (Y30 -Y3M ) R2 N C (Y30 -Y3M ) R2 N C (Y30 -Y3M ) R2 N
1.03 −0.81 −10.26 4.11 4.35 −1.08
1 0.0002 62 0.01 48 0.0002 122
(0.95) (0.89) (0.62) (0.53) (0.78) (0.85)
−0.25 −0.87 −13.06 4.96 2.07 −0.28
2 0.001 61 0.02 47 0.000 121
(0.98) (0.85) (0.47) (0.39) (0.88) (0.95)
−1.98 −0.33 −17.59 6.28 −0.74 0.66
3 0.0002 60 0.06 46 0.0003 120
(0.84) (0.94) (0.26) (0.22) (0.95) (0.89)
−5.05 0.45 −16.06 5.65 −3.24 1.70
4 0.001 59 0.06 45 0.003 119
(0.51) (0.90) (0.30) (0.27) (0.77) (0.68)
J. Risk Financial Manag. 2021, 14, 585 20 of 39

Table 8. Cont.

Panel B
Forecast
Oil (1986:01–2003:12) Silver (1986:01–2003:12) Gold (1986:01–2003:12)
Horizon
h C (Y30 -Y3M ) R2 N C (Y30 -Y3M ) R2 N C (Y30 -Y3M ) R2 N
13.64 −3.74 −8.02 4.36 * −5.61 3.29 *
1 0.01 71 0.03 71 0.03 70
(0.31) (0.42) (0.14) (0.07) (0.19) (0.05)
10.91 −2.46 −10.57 ** 6.02 ** −5.54 3.29 **
2 0.01 71 0.11 71 0.08 71
(0.35) (0.52) (0.04) (0.03) (0.11) (0.02)
10.37 −2.29 −10.29 ** 5.64 *** −5.37 * 2.95 ***
3 0.01 71 0.14 71 0.11 71
(0.32) (0.52) (0.02) (0.01) (0.06) (0.01)
8.64 −1.44 −9.87 *** 5.44 *** −4.89* 2.65 **
4 0.005 71 0.18 71 0.11 71
(0.38) (0.68) (0.01) (0.004) (0.08) (0.02)
Forecast
Platinum (1986:04–2003:12) Palladium (1986:04–2003:12) Zinc (1997:08–2003:12)
Horizon
h C (Y30 -Y3M ) R2 N C (Y30 -Y3M ) R2 N C (Y30 -Y3M ) R2 N
−1.58 2.69 13.62 −4.86 −18.45 7.06 *
1 0.01 71 0.01 71 0.10 25
(0.85) (0.34) (0.40) (0.43) (0.11) (0.09)
−4.02 3.98 6.92 −1.14 −17.37 * 7.92 *
2 0.05 71 0.001 71 0.20 25
(0.57) (0.10) (0.65) (0.85) (0.09) (0.05)
−5.57 4.38 ** 1.89 0.97 −16.08 * 7.34 **
3 0.10 70 0.001 71 0.25 25
(0.39) (0.04) (0.90) (0.86) (0.09) (0.04)
−6.05 4.60 ** −0.58 2.13 −15.39 * 7.45 **
4 0.14 71 0.01 71 0.33 25
(0.30) (0.02) (0.97) (0.69) (0.08) (0.02)
Forecast
Natural gas (1990:05–2003:12)
Horizon
h C (Y30 -Y3M ) R2 N
10.84 −0.20
1 0.000 54
(0.71) (0.98)
1.57 3.48
2 0.01 54
(0.95) (0.67)
−3.79 5.79
3 0.02 54
(0.87) (0.43)
−6.18 7.15
4 0.06 54
(0.75) (0.24)
Panel C
Forecast
Oil (2004:01–2020:12) Silver (2004:01–2020:12) Gold (2004:01–2020:12)
Horizon
h C (Y30 -Y3M ) R2 N C (Y30 -Y3M ) R2 N C (Y30 -Y3M ) R2 N
5.77 −1.43 17.50 * −3.79 16.24 *** −3.14
1 0.0004 68 0.01 68 0.03 68
(0.75) (0.82) (0.09) (0.42) (0.01) (0.18)
5.29 −1.57 18.60 ** −4.66 16.74 *** −3.30
2 0.001 67 0.02 67 0.05 67
(0.73) (0.78) (0.04) (0.28) (0.003) (0.16)
3.24 −0.89 15.15 * −3.26 16.34 *** −3.05
3 0.001 66 0.02 66 0.06 66
(0.82) (0.87) (0.07) (0.42) (0.002) (0.16)
−1.50 0.41 11.55 −2.02 16.08 *** −2.98
4 0.0002 65 0.01 65 0.07 65
(0.91) (0.93) (0.15) (0.60) (0.001) (0.13)
J. Risk Financial Manag. 2021, 14, 585 21 of 39

Table 8. Cont.

Panel C
Forecast
Platinum (2004:01–2020:12) Palladium (2004:01–2020:12) Zinc (2004:01–2020:12)
Horizon
h C (Y30 -Y3M ) R2 N C (Y30 -Y3M ) R2 N C (Y30 -Y3M ) R2 N
11.93 −4.47 27.18 *** −5.38 12.91 −2.90
1 0.01 68 0.01 68 0.004 68
(0.14) (0.21) (0.01) (0.33) (0.54) (0.67)
12.74 −5.13 28.83 *** −6.52 6.80 −0.66
2 0.04 67 0.03 67 0.0004 67
(0.11) (0.16) (0.003) (0.22) (0.73) (0.92)
9.95 −3.96 24.38 *** −4.56 1.45 1.52
3 0.04 66 0.03 66 0.003 66
(0.16) (0.19) (0.01) (0.34) (0.94) (0.79)
7.52 −3.09 21.37 ** −3.16 −2.77 3.11
4 0.04 65 0.02 65 0.02 65
(0.27) (0.24) (0.04) (0.47) (0.88) (0.55)
Forecast
Natural gas (2004:01–2020:12)
Horizon
h C (Y30 -Y3M ) R2 N
−1.68 −1.37
1 0.000 68
(0.92) (0.81)
1.87 −2.89
2 0.004 67
(0.91) (0.61)
1.14 −3.01
3 0.01 66
(0.93) (0.55)
−1.64 −2.16
4 0.01 65
(0.89) (0.62)
Notes: The forecast horizon (h) is in quarters. Y30 -Y3M denotes the yield spread calculated as the difference between the yield rates on
10-year and 3-month government bonds. The table reports the estimation results of Equation (1) with the Newey and West (1987) procedure.
The sample period appears separately for each commodity. Figures in parentheses denote estimated standard errors. ***, ** and * denote
statistical significance at the 1%, 5%, and 10% levels, respectively.

Table 9. Estimation results of Equation (1) with the Y30 -Y1 indicator. Panel A: entire sample. Panel C: sample period
2004–2020.

Panel A
Forecast
Oil (1986:01–2020:12) Silver (1986:01–2020:12) Gold (1986:01–2020:12)
Horizon
h C (Y30 -Y1 ) R2 N C (Y30 -Y1 ) R2 N C (Y30 -Y1 ) R2 N
9.77 −2.90 1.75 1.55 3.58 0.74
1 0.002 139 0.002 139 0.001 138
(0.28) (0.43) (0.79) (0.63) (0.44) (0.69)
7.95 −2.18 1.51 1.63 3.40 0.83
2 0.003 138 0.004 138 0.003 138
(0.34) (0.52) (0.81) (0.59) (0.45) (0.65)
7.13 −1.90 0.45 1.98 3.09 0.86
3 0.004 137 0.01 137 0.005 137
(0.36) (0.57) (0.94) (0.49) (0.47) (0.62)
4.93 −1.15 −0.66 2.37 3.07 0.78
4 0.002 136 0.02 136 0.01 136
(0.51) (0.72) (0.90) (0.39) (0.46) (0.64)
Forecast
Platinum (1986:04–2020:12) Palladium (1986:04–2020:12) Zinc (1997:08–2020:12)
Horizon
h C (Y30 -Y1 ) R2 N C (Y30 -Y1 ) R2 N C (Y30 -Y1 ) R2 N
3.68 −0.50 17.84 * −4.63 1.52 0.81
1 0.0002 139 0.01 139 0.0004 93
(0.52) (0.84) (0.07) (0.32) (0.91) (0.87)
3.41 −0.50 16.10 * −3.75 −1.001 2.08
2 0.0005 138 0.01 138 0.005 92
(0.53) (0.84) (0.09) (0.40) (0.94) (0.64)
2.09 −0.03 12.74 −2.18 −3.21 3.09
3 0.000 136 0.01 137 0.01 91
(0.69) (0.99) (0.18) (0.60) (0.80) (0.45)
0.96 0.46 10.23 −0.85 −5.19 4.02
4 0.001 136 0.001 136 0.03 90
(0.84) (0.83) (0.27) (0.83) (0.66) (0.28)
J. Risk Financial Manag. 2021, 14, 585 22 of 39

Table 9. Cont.

Panel A
Forecast
Ethanol (2005:06–2020:12) Coal (2009:01–2020:12) Natural gas (1990:05–2020:12)
Horizon
h C (Y30 -Y1 ) R2 N C (Y30 -Y1 ) R2 N C (Y30 -Y1 ) R2 N
−0.37 −0.20 −7.99 3.47 4.98 −1.52
1 0.000 62 0.01 48 0.0005 122
(0.98) (0.97) (0.70) (0.61) (0.73) (0.79)
−1.43 −0.38 −10.36 4.21 2.60 −0.58
2 0.000 61 0.02 47 0.0001 121
(0.90) (0.94) (0.57) (0.50) (0.84) (0.91)
−1.98 −0.36 −14.18 5.35 0.99 −0.10
3 0.0002 60 0.04 46 0.000 120
(0.83) (0.93) (0.38) (0.34) (0.93) (0.98)
−4.89 0.41 −12.75 4.74 0.04 −1.07 0.85
4 0.000 59 45 0.001 119
(0.51) (0.91) (0.04) (0.39) (0.92) (0.84)
Panel B
Forecast
Oil (1986:01–2003:12) Silver (1986:01–2003:12) Gold (1986:01–2003:12)
Horizon
h C (Y30 -Y1 ) R2 N C (Y30 -Y1 ) R2 N C (Y30 -Y1 ) R2 N
13.74 −4.59 −8.56 * 5.60 ** −5.65 4.02 **
1 0.01 71 0.04 71 0.04 70
(0.24) (0.30) (0.09) (0.04) (0.16) (0.03)
10.92 −2.99 −10.78 ** 7.42 ** −6.00 * 4.26 ***
2 0.01 71 0.14 71 0.12 71
(0.30) (0.44) (0.03) (0.02) (0.07) (0.005)
−10.63
9.39 −2.20 7.04 *** −6.14 ** 4.03 ***
3 0.01 71 *** 0.19 71 0.17 71
(0.32) (0.54) (0.003) (0.03) (0.001)
(0.01)
−10.29
7.91 −1.31 6.85 *** −5.87 ** 3.79 ***
4 0.004 71 *** 0.24 71 0.19 71
(0.37) (0.71) (0.001) (0.03) (0.001)
(0.005)
Forecast
Platinum (1986:04–2003:12) Palladium (1986:04–2003:12) Zinc (1997:08–2003:12)
Horizon
h C (Y30 -Y1 ) R2 N C (Y30 -Y1 ) R2 N C (Y30 -Y1 ) R2 N
−0.97 2.89 15.75 −7.15 −17.19 7.11
1 0.01 71 0.02 71 0.09 25
(0.90) (0.35) (0.27) (0.30) (0.15) (0.12)
−3.10 4.27 9.86 −3.11 −16.06 8.03 *
2 0.05 71 0.01 71 0.19 25
(0.65) (0.12) (0.48) (0.64) (0.12) (0.07)
−4.49 4.66 * 5.65 −1.04 −15.05 7.54 **
3 0.10 70 0.001 71 0.24 25
(0.47) (0.05) (0.68) (0.86) (0.11) (0.04)
−5.10 5.02 ** 2.85 0.56 −14.51* 7.75 **
4 0.15 71 0.0004 71 0.32 25
(0.36) (0.02) (0.83) (0.92) (0.09) (0.02)
Forecast
Natural gas (1990:05–2003:12)
Horizon
h C (Y30 -Y1 ) R2 N
12.74 −1.23
1 0.0002 54
(0.62) (0.90)
4.83 2.35
2 0.002 54
(0.83) (0.77)
0.74 4.38
3 0.01 54
(0.97) (0.54)
−1.10 5.68
4 0.03 54
(0.95) (0.34)
J. Risk Financial Manag. 2021, 14, 585 23 of 39

Table 9. Cont.

Panel C
Forecast
Oil (2004:01–2020:12) Silver (2004:01–2020:12) Gold (2004:01–2020:12)
Horizon
h C (Y30 -Y1 ) R2 N C (Y30 -Y1 ) R2 N C (Y30 -Y1 ) R2 N
4.55 −0.99 15.55 −3.23 15.62 *** −3.15
1 0.0002 68 0.01 68 0.02 68
(0.79) (0.87) (0.13) (0.52) (0.01) (0.23)
3.67 −0.97 17.50 * −4.59 16.35 *** −3.44
2 0.0004 67 0.02 67 0.05 67
(0.80) (0.86) (0.06) (0.33) (0.003) (0.18)
3.55 −1.12 15.24 * −3.62 16.21 *** −3.29
3 0.001 66 0.02 66 0.06 66
(0.79) (0.83) (0.07) (0.41) (0.002) (0.17)
−0.10 −0.19 12.42 −2.61 16.13 *** −3.29
4 0.000 65 0.02 65 0.08 65
(0.99) (0.97) (0.12) (0.53) (0.001) (0.13)
Forecast
Platinum (2004:01–2020:12) Palladium (2004:01–2020:12) Zinc (2004:01–2020:12)
Horizon
h C (Y30 -Y1 ) R2 N C (Y30 -Y1 ) R2 N C (Y30 -Y1 ) R2 N
8.85 −3.45 22.91 ** −3.90 12.24 −2.86
1 0.01 68 0.01 68 0.004 68
(0.27) (0.31) (0.03) (0.48) (0.55) (0.69)
10.67 −4.65 25.87 *** −5.76 7.79 −1.18
2 0.03 67 0.02 67 0.001 67
(0.15) (0.18) (0.01) (0.28) (0.69) (0.86)
9.57 −4.17 23.83 *** −4.74 3.82 0.57
3 0.04 66 0.03 66 0.0004 66
(0.17) (0.18) (0.01) (0.33) (0.84) (0.93)
7.76 −3.51 21.88 ** −3.70 0.47 1.92
4 0.05 65 0.02 65 0.01 65
(0.23) (0.19) (0.03) (0.41) (0.98) (0.73)
Forecast
Natural Gas (2004:01–2020:12)
Horizon
h C (Y30 -Y1 ) R2 N
−3.02 −0.87
1 0.0002 68
(0.86) (0.88)
−0.63 −2.01
2 0.002 67
(0.97) (0.71)
−0.26 −2.66
3 0.01 66
(0.98) (0.59)
−2.82 −1.83
4 0.004 65
(0.81) (0.68)
Notes: The forecast horizon (h) is in quarters. Y30 -Y1 denotes the yield spread calculated as the difference between the yield rates on
10-year and 1-year government bonds. The table reports the estimation results of Equation (1) with the Newey and West (1987) procedure.
The sample period appears separately for each commodity. Figures in parentheses denote estimated standard errors. ***, ** and * denote
statistical significance at the 1%, 5%, and 10% levels, respectively.

Table 10. Estimation results of Equation (1) with the Y30 -Y2 indicator. Panel A: entire sample. Panel B: sample period
1986–2003. Panel C: sample period 2004–2020.

Panel A
Forecast
Oil (1986:01–2020:12) Silver (1986:01–2020:12) Gold (1986:01–2020:12)
Horizon
h C (Y30 -Y2 ) R2 N C (Y30 -Y2 ) R2 N C (Y30 -Y2 ) R2 N
10.05 −3.63 0.99 2.32 2.96 1.28
1 0.003 139 0.003 139 0.003 138
(0.24) (0.37) (0.87) (0.53) (0.50) (0.55)
7.63 −2.40 0.94 2.30 2.85 1.33
2 0.003 138 0.01 138 0.01 138
(0.33) (0.52) (0.87) (0.51) (0.49) (0.51)
6.90 −2.12 0.30 2.45 2.52 1.38
3 0.004 137 0.01 137 0.01 137
(0.34) (0.56) (0.96) (0.46) (0.53) (0.47)
5.09 −1.47 −0.39 2.66 2.51 1.29
4 0.003 136 0.02 136 0.01 136
(0.46) (0.68) (0.94) (0.40) (0.51) (0.48)
J. Risk Financial Manag. 2021, 14, 585 24 of 39

Table 10. Cont.

Panel A
Forecast
Platinum (1986:04–2020:12) Palladium (1986:04–2020:12) Zinc (1997:08–2020:12)
Horizon
h C (Y30 -Y2 ) R2 N C (Y30 -Y2 ) R2 N C (Y30 -Y2 ) R2 N
3.25 −0.33 16.08 * −4.42 1.81 0.75
1 0.000 139 0.01 139 0.0003 93
(0.56) (0.91) (0.09) (0.40) (0.89) (0.89)
3.20 −0.47 15.32 * −4.00 −0.29 1.95
2 0.0003 138 0.01 138 0.003 92
(0.54) (0.87) (0.09) (0.42) (0.98) (0.69)
2.49 −0.28 13.01 −2.77 −2.13 2.90
3 0.0002 136 0.01 137 0.01 91
(0.62) (0.92) (0.15) (0.55) (0.86) (0.52)
1.64 0.12 10.71 −1.32 −3.72 3.73
4 0.0001 136 0.002 136 0.02 90
(0.72) (0.96) (0.23) (0.76) (0.75) (0.36)
Forecast
Ethanol (2005:06–2020:12) Coal (2009:01–2020:12) Natural Gas (1990:05–2020:12)
Horizon
h C (Y30 -Y2 ) R2 N C (Y30 -Y2 ) R2 N C (Y30 -Y2 ) R2 N
−1.17 0.20 −7.47 3.59 6.34 −2.57
1 0.000 62 0.01 48 0.001 122
(0.94) (0.98) (0.71) (0.63) (0.65) (0.68)
−2.25 0.01 −8.79 3.94 4.53 −1.80
2 0.000 61 0.01 47 0.001 121
(0.84) (0.99) (0.64) (0.58) (0.72) (0.75)
−1.75 −0.51 −12.05 4.97 3.33 −1.46
3 0.0003 60 0.03 46 0.001 120
(0.85) (0.92) (0.48) (0.44) (0.77) (0.78)
−4.24 0.12 −10.28 4.16 1.17 −0.28
4 0.000 59 0.03 45 0.000 119
(0.58) (0.98) (0.54) (0.52) (0.91) (0.95)
Panel B
Forecast
Oil (1986:01–2003:12) Silver (1986:01–2003:12) Gold (1986:01–2003:12)
Horizon
h C (Y30 -Y2 ) R2 N C (Y30 -Y2 ) R2 N C (Y30 -Y2 ) R2 N
13.76 −6.08 −7.65 6.70 ** −5.00 4.81 **
1 0.01 71 0.05 71 0.04 70
(0.20) (0.22) (0.11) (0.04) (0.20) (0.02)
10.50 −3.63 −9.78 ** 9.03 *** −5.45 * 5.21 ***
2 0.01 71 0.15 71 0.13 71
(0.28) (0.41) (0.03) (0.01) (0.08) (0.002)
8.50 −2.22 −9.99 *** 8.81 *** −5.83 ** 5.09 ***
3 0.01 71 0.23 71 0.21 71
(0.33) (0.58) (0.01) (0.001) (0.03) (0.0001)
6.63 −0.73 −9.64 *** 8.55 *** −5.68 ** 4.87 ***
4 0.001 71 0.29 71 0.23 71
(0.41) (0.86) (0.01) (0.000) (0.02) (0.00)
Forecast
Platinum (1986:04–2003:12) Palladium (1986:04–2003:12) Zinc (1997:08–2003:12)
Horizon
h C (Y30 -Y2 ) R2 N C (Y30 -Y2 ) R2 N C (Y30 -Y2 ) R2 N
−0.44 3.41 14.97 −8.85 −15.13 7.27
1 0.01 71 0.02 71 0.08 25
(0.95) (0.37) (0.25) (0.27) (0.20) (0.16)
−2.22 4.96 10.69 −4.78 −14.62 8.84 *
2 0.06 71 0.01 71 0.19 25
(0.74) (0.14) (0.40) (0.54) (0.14) (0.06)
−3.44 5.33 * 7.10 −2.52 −14.04 8.85 **
3 0.10 70 0.005 71 0.26 25
(0.57) (0.07) (0.56) (0.72) (0.12) (0.02)
−3.99 5.77 ** 4.21 −0.32 −13.54 * 8.84 ***
4 0.14 71 0.000 71 0.34 25
(0.46) (0.02) (0.72) (0.96) (0.09) (0.004)
J. Risk Financial Manag. 2021, 14, 585 25 of 39

Table 10. Cont.

Panel B
Forecast
Natural Gas (1990:05–2003:12)
Horizon
h C (Y30 -Y2 ) R2 N
13.81 −2.31
1 0.001 54
(0.55) (0.84)
8.26 0.71
2 0.0001 54
(0.69) (0.94)
4.46 3.12
3 0.005 54
(0.81) (0.69)
2.21 5.08
4 0.02 54
(0.89) (0.43)
Panel C
Forecast
Oil (2004:01–2020:12) Silver (2004:01–2020:12) Gold (2004:01–2020:12)
Horizon
h C (Y30 -Y2 ) R2 N C (Y30 -Y2 ) R2 N C (Y30 -Y2 ) R2 N
4.49 −1.07 15.00 −3.28 15.76 *** −3.55
1 0.000 68 0.004 68 0.02 68
(0.79) (0.87) (0.17) (0.59) (0.01) (0.25)
2.74 −0.59 17.44 * −5.04 16.72 *** −3.98
2 0.0001 67 0.02 67 0.05 67
(0.85) (0.92) (0.07) (0.36) (0.004) (0.18)
3.61 −1.27 16.49 * −4.63 16.89 *** −3.98
3 0.001 66 0.03 66 0.07 66
(0.78) (0.82) (0.06) (0.37) (0.002) (0.15)
1.10 −0.82 14.29 * −3.83 16.95 *** −4.05
4 0.001 65 0.03 65 0.10 65
(0.93) (0.88) (0.07) (0.42) (0.00) (0.11)
Forecast
Platinum (2004:01–2020:12) Palladium (2004:01–2020:12) Zinc (2004:01–2020:12)
Horizon
h C (Y30 -Y2 ) R2 N C (Y30 -Y2 ) R2 N C (Y30 -Y2 ) R2 N
7.76 −3.24 21.83 * −3.74 13.28 −3.69
1 0.01 68 0.004 68 0.01 68
(0.35) (0.37) (0.05) (0.54) (0.52) (0.64)
9.88 −4.73 25.47 *** −6.15 9.62 −2.25
2 0.02 67 0.02 67 0.003 67
(0.19) (0.19) (0.01) (0.28) (0.63) (0.76)
10.02 −4.84 25.20 *** −5.94 6.23 −0.61
3 0.04 66 0.03 66 0.0003 66
(0.16) (0.16) (0.01) (0.27) (0.75) (0.93)
8.70 −4.35 23.84 *** −5.08 3.33 0.67
4 0.05 65 0.03 65 0.001 65
(0.17) (0.14) (0.01) (0.29) (0.86) (0.92)
Forecast
Natural Gas (2004:01–2020:12)
Horizon
h C (Y30 -Y2 ) R2 N
−3.61 −0.66
1 0.000 68
(0.84) (0.92)
−1.65 −1.69
2 0.001 67
(0.92) (0.78)
−0.53 −2.79
3 0.005 66
(0.97) (0.62)
−3.15 −1.85
4 0.003 65
(0.79) (0.71)
Notes: The forecast horizon (h) is in quarters. Y30 -Y2 denotes the yield spread calculated as the difference between the yield rates on
10-year and 2-year government bonds. The table reports the estimation results of Equation (1) with the Newey and West (1987) procedure.
The sample period appears separately for each commodity. Figures in parentheses denote estimated standard errors. ***, ** and * denote
statistical significance at the 1%, 5%, and 10% levels, respectively.
J. Risk Financial Manag. 2021, 14, 585 26 of 39

6.2. Commodity Prices following Periods of Non-Positive Yield Spreads

What happens to the price of commodities in the months following downward-sloped
or flat yield curves? To answer this question, we track the evolution in the price of
the sampled commodities throughout the sample period. The period between 1986 and
December 2020 witnessed eight periods during which the yield spread (10-year minus
1-year yields) was negative or equal to zero. The left-hand side of Table 11 lists the dates
when the yield spread became non-positive. For each commodity, we calculate the returns
accumulated 1, 2, 3, 4, 5 and 6 quarters following these non-positive yield spread periods.

Table 11. Commodity prices in the periods following non-positive yield spread. Panel A: oil. Panel B: silver. Panel C: gold.
Panel D: platinum. Panel E: palladium. Panel F: zinc. Panel G: ethanol. Panel H: coal. Panel I: natural gas.

Panel A
Periods Associated with
Return Return Return Return Return Return
Non-Positive
1Q Later 2 Qs. Later 3Qs. Later 4Qs. Later 5Qs. Later 6Qs. Later
Yield Spreads
Start End
25/01/1989 30/06/1989 −0.69% 7.65% 0.05% −15.79% 82.98% 40.31%
04/08/1989 13/10/1989 10.72% −14.89% −12.11% 90% 47.34% 2.82%
17/03/2000 28/04/2000 9.48% 27.2% 12.9% 9.83% 4.97% −13.95%
27/12/2005 29/03/2006 10.64% −5.33% −8.13% −0.63% 6.37% 22.89%
05/06/2006 05/06/2007 15.42% 33.35% 59.3% 94.77% 61.91% −37.8%
20/07/2007 08/08/2007 32.31% 27.19% 71.43% 59.67% −15.4% −45.17%
23/05/2019 03/06/2019 1.3% 5.35% −11.4% −30.87% −22.31% −14.29%
05/08/2019 08/10/2019 13.26% −52.33% −22.29% −21.74%
Average 11.56% 3.52% 11.22% 23.15% 23.69% −6.46%
Panel B
Periods Associated with
Return Return Return Return Return Return
Non-Positive
1 Q Later 2 Qs. Later 3Qs. Later 4Qs. Later 5Qs. Later 6Qs. Later
Yield Spreads
Start End
25/01/1989 30/06/1989 3.01% 2.29% −3.39% −5.2% −8.63% −17.61%
04/08/1989 13/10/1989 4.82% −0.62% −4.51% −15.27% −17.18% −22.41%
17/03/2000 28/04/2000 0.81% −3.63% −2.76% −12.16% −15.28% −14.46%
27/12/2005 29/03/2006 −6.68% 4.22% 16.82% 20.47% 11.56% 25.71%
05/06/2006 05/06/2007 −11.74% 3.32% 49.76% 24.31% −11.35% −31.94%
20/07/2007 08/08/2007 17.81% 29.92% 27.52% 16.4% −24.35% −2.58%
23/05/2019 03/06/2019 29.44% 16.18% 16.21% 23.88% 81.38% 63.41%
05/08/2019 08/10/2019 2.64% −14.1% 7.8% 34.89%
Average 5.01% 4.7% 13.43% 10.92% 2.31% 0.02%
Panel C
Periods Associated with
Return Return Return Return Return Return
Non-Positive
1 Q Later 2 Qs. Later 3Qs. Later 4Qs. Later 5Qs. Later 6Qs. Later
Yield Spreads
Start End
25/01/1989 30/06/1989 −2.68% 7.39% −1.88% −4.61% 2.46% 5.01%
04/08/1989 13/10/1989 14.15% 3.47% 0.66% 6.66% 10.54% −0.55%
17/03/2000 28/04/2000 1.09% −3.75% −4.33% −3.79% −2.66% 1.6%
27/12/2005 29/03/2006 2.72% 4.41% 11.29% 15.38% 13.54% 29.57%
05/06/2006 05/06/2007 2.2% 19.1% 47.63% 30.2% 19.3% 12.08%
20/07/2007 08/08/2007 24.17% 36.16% 30.78% 27.18% 8.85% 32.31%
23/05/2019 03/06/2019 17.15% 11.76% 24.32% 30.43% 45.93% 38.87%
05/08/2019 08/10/2019 4.21% 11.23% 21.6% 26.14%
Average 7.88% 11.22% 16.26% 15.95% 13.99% 16.98%
J. Risk Financial Manag. 2021, 14, 585 27 of 39

Table 11. Cont.

Panel D
Periods Associated with
Return Return Return Return Return Return
Non-Positive
1 Q Later 2 Qs. Later 3Qs. Later 4Qs. Later 5Qs. Later 6Qs. Later
Yield Spreads
Start End
25/01/1989 30/06/1989 −0.42% −1.73% −4.32% −2.18% −13.93% −17.58%
04/08/1989 13/10/1989 3.49% −0.83% −0.99% −13.79% −13.63% −17.16%
17/03/2000 28/04/2000 13.43% 13.6% 19.99% 18.93% 1% −15.14%
27/12/2005 29/03/2006 11.96% 5.97% 5.79% 15.39% 18.77% 29.56%
05/06/2006 05/06/2007 −1.99% 13.05% 75.25% 54.95% 5.31% −39.39%
20/07/2007 08/08/2007 14.14% 45.92% 58.18% 20.8% −34.01% −22.86%
23/05/2019 03/06/2019 16.41% 11.05% 5.9% 5.82% 8.37% 26.52%
05/08/2019 08/10/2019 8.32% −17.59% −1.28% −3.06%
Average 8.17% 8.68% 19.82% 12.11% −4.02% −8.01%
Panel E
Periods Associated with
Return Return Return Return Return Return
Non-Positive
1 Q Later 2 Qs. Later 3Qs. Later 4Qs. Later 5Qs. Later 6Qs. Later
Yield Spreads
Start End
25/01/1989 30/06/1989 −8.94% −13.75% −17.41% −25.11% −38.73% −48.28%
04/08/1989 13/10/1989 −0.93% −6.91% −16.05% −32.23% −35.82% −29.44%
17/03/2000 28/04/2000 24.56% 21.82% 69.84% 10.91% −26.72% −44.09%
27/12/2005 29/03/2006 −6.35% −5.45% 1.15% 6.38% 10.12% 5.17%
05/06/2006 05/06/2007 −9.21% −4.44% 52.74% 15.44% −26.12% −55.75%
20/07/2007 08/08/2007 3.72% 21.13% 19.77% −9.12% −38.45% −43.44%
23/05/2019 03/06/2019 16.71% 39.16% 83.09% 48.74% 74.85% 75.24%
05/08/2019 08/10/2019 24.88% 26.93% 17.42% 45.57%
Average 5.55% 9.81% 26.32% 7.57% −11.55% −20.08%
Panel F
Periods Associated with
Return Return Return Return Return Return
Non-Positive
1 Q Later 2 Qs. Later 3Qs. Later 4Qs. Later 5Qs. Later 6Qs. Later
Yield Spreads
Start End
17/03/2000 28/04/2000 0.02% −9.09% −9.56% −17.29% −27% −34.66%
27/12/2005 29/03/2006 19.53% 27.26% 64.03% 24.16% 28% 17.24%
05/06/2006 05/06/2007 −25.82% −36.91% −26.28% −49.13% −54.42% −72.24%
20/07/2007 08/08/2007 −20.16% −30.17% −37.26% −52.14% −69.52% −66.48%
23/05/2019 03/06/2019 −14.05% −14.01% −23.99% −21.75% −3.41% 6.25%
05/08/2019 08/10/2019 4.25% −17.65% −8.67% 1.14%
Average −6.04% −13.43% −6.96% −19.17% −25.27% −29.98%
Panel G
Periods Associated with
Return Return Return Return Return Return
Non-Positive
1 Q Later 2 Qs. Later 3Qs. Later 4Qs. Later 5Qs. Later 6Qs. Later
Yield Spreads
Start End
27/12/2005 29/03/2006 35.07% −29.26% −0.08% −9.02% −21.84% −37.88%
05/06/2006 05/06/2007 −22.17% −7.31% 11.04% 10.47% 2.36% −34.06%
20/07/2007 08/08/2007 0.54% 14.36% 40.65% 10.57% −6.67% −12.47%
23/05/2019 03/06/2019 −12.49% −2.91% −16.66% −23.33% −9.78% −7.47%
05/08/2019 08/10/2019 −7.42% −38.07% −5.69% −3.26%
Average −1.29% −12.64% 5.85% −2.91% −8.98% −22.97%
J. Risk Financial Manag. 2021, 14, 585 28 of 39

Table 11. Cont.

Panel H
Periods Associated with
Return Return Return Return Return Return
Non-Positive
1 Q Later 2 Qs. Later 3Qs. Later 4Qs. Later 5Qs. Later 6Qs. Later
Yield Spreads
Start End
23/05/2019 03/06/2019 −9.22% −7.99% −9.62% −24.16% −32.15% 0.41%
05/08/2019 08/10/2019 0.3% −6.35% −21.45% −13.3%
Average −4.46% −7.17% −15.54% −18.73% −32.15% 0.41%
Panel I
Periods Associated with
Return Return Return Return Return Return
Non-Positive
1 Q Later 2 Qs. Later 3Qs. Later 4Qs. Later 5Qs. Later 6Qs. Later
Yield Spreads
Start End
17/03/2000 28/04/2000 22.41% 44.57% 100.35% 54.95% 0.83% 1.94%
27/12/2005 29/03/2006 −15.18% −22.3% −12.91% 5.2% −6.36% −5.02%
05/06/2006 05/06/2007 −28.01% −10.9% 20.8% 55.25% −7.63% −28.79%
20/07/2007 08/08/2007 24% 33.46% 81.08% 32.6% 8.63% −22.72%
23/05/2019 03/06/2019 −1.87% 1.58% −25.09% −26.05% 3.5% 4.33%
05/08/2019 08/10/2019 −6.42% −22.07% −20.28% 14.82%
Average −0.85% 4.06% 23.99% 22.79% −0.21% −10.05%
Notes: the tables report the commodity returns accumulated after 1, 2, 3, 4, 5 and 6 quarters following downward-sloped or flat yield
curves.

Oil, silver, gold, platinum, palladium, and natural gas prices surged strongly in the
quarters following the periods associated with equality in long and short-term Treasury
yields. For example, tracking the prices of these commodities three quarters after an end in
the zero slope in the bond term structure reveals significant positive returns on average
(oil 11.22%; silver 13.43%; gold 16.26%; platinum 19.82%; palladium 26.32%; ethanol 5.85%;
natural gas 24%). In contrast, coal and zinc prices present a mixed and inconclusive picture
with a tendency to negative returns. This finding emphasizes that investors should note
that flat or downward-sloped yield curves seem to be reasonable points at which to take
long positions in several commodities that they plan to hold for a relatively long period of
time.
Our findings are even more pronounced if we consider the recent relatively flat
yield curve observed during the last week in February 2020 due to the outbreak of the
coronavirus. However, we did not include the findings in the table because the difference
between the 10-year and 1-year bond interest rates was 0.03% (0.0003). While quite small, it
is not a non-positive yield. In addition, prices recovered sharply after two to four quarters.
Nevertheless, the findings in Table A1 in the online Appendix A lend support to our
conjecture.

7. Conclusions
We investigated an important, yet barely discussed, issue: Can yield spreads forecast
future innovations in the commodity market? If so, is this long-term correlation stable over
time? Despite the extensive research linking economic real activity to lagged yield spreads,
the predictive ability of the yield curve has not been proven with regard to commodities
often used in as raw materials.
Our findings can be summarized as follows. First, the prediction ability of the yield
curve is evident mainly in the period before the financialization of commodities era, but is
absent between 2004 and 2020. Second, structural break tests confirm the changes in the
correlation between the six yield spreads proposed and future commodity prices. Third,
the findings of the dynamic conditional correlation confirm the time-varying nature of the
J. Risk Financial Manag. 2021, 14, 585 29 of 39

yield spread in predicting the future evolution in commodity prices. One explanation might
be the increased flow of money into the commodity market and the increased correlation
between it and equity markets. These changes disconnected the prices of commodities
from the economic cycle.
The structural breaks and the fading correlation between the variables of interest are
critical for those involved in risk management and investment diversification. Furthermore,
our results may be useful for policy makers who must make decisions about policies
to target and control inflation. Future research can extend the standing literature by
addressing the interplay between the shape of the term structure and future evolution of
asset prices in the wake of pandemic outbreaks and the massive monetary intervention
conducted by central banks under severe economic conditions.

Author Contributions: Conceptualization, Y.I.-B. and M.Q.; methodology, Y.I.-B. and M.Q.; software,
Y.I.-B. and M.Q.; validation, Y.I.-B. and M.Q.; formal analysis, Y.I.-B. and M.Q.; investigation, Y.I.-B.
and M.Q.; resources, M.Q.; data curation, Y.I.-B. and M.Q.; writing—original draft preparation, Y.I.-B.
and M.Q.; writing—review and editing, Y.I.-B. and M.Q.; visualization, Y.I.-B. and M.Q.; supervision,
M.Q.; project administration, M.Q. All authors have read and agreed to the published version of the
manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.

Appendix A
Table A1. Evolution in the commodity prices following the relatively flat curve witnessed in February 2020.

Accumulated Returns after:

2 3 4 5 6 7 8 9 10
1 Month
Months Months Months Months Months Months Months Months Months
Oil −50.9% −66.1% −33.4% −22.4% −17.3% −13.1% −19.3% −22.7% −8.4% −1.6%
Silver −18.4% −16.1% −2.9% −1.6% 25.4% 44.4% 26.5% 34.2% 28.4% 42.6%
Gold −0.8% 4.7% 5.4% 7.0% 15.2% 16.1% 12.8% 15.5% 9.6% 14.5%
PLTNM −20.0% −17.0% −4.9% −13.9% 1.7% 0.2% −9.7% −6.4% 4.0% 11.9%
PLDM −14.0% −25.0% −25.3% −31.1% −13.4% −18.0% −16.6% −10.5% −11.5% −10.8%
Zinc −9.4% −7.1% −1.6% 1.3% 9.9% 22.2% 16.9% 25.4% 35.9% 41.7%
Ethanol −29.6% −27.5% −14.3% −12.0% −10.9% −0.4% 1.6% 21.5% 6.2% 2.3%
Coal −2.3% −10.0% −23.5% −21.6% −23.4% −26.7% −22.2% −14.6% −6.1% 17.8%
Nat.Gas −10.2% −5.5% −6.3% −19.8% −2.1% 34.8% 15.8% 63.7% 56.8% 46.2%
Notes: With the outbreak of the COVID−19 pandemic in February 2020, the U.S. 1-year yield was 1.43% while the 10-year was 1.46%. That
is, the yield spread was 0.03%. The table reports the evolution in the prices of commodities in the few months following this relatively flat
curve. Though we discuss one case, the overall picture is clear and shows that the current yield spread is a relatively good predictor of the
future evolution in commodity prices.

Table A2. Estimation results of model 5 with the Y10-Y3M indicator. Panel A: entire sample. Panel B: sample period
2004–2020.

Panel A
Forecast
Horizon Oil (1986:01–2020:12) Silver (1986:01–2020:12)

(Y10 - ∆SP ∆EX ∆IP ∆EPU (Y10 - ∆SP ∆EX ∆IP ∆EPU
h C
Y3M ) R2 N C
Y3M ) R2 N

5.92 −0.79 −0.55 9.32 −14.07b 0.11 0.03 139 1.67 1.06 −0.85 −10.29 −4.90 0.06 −0.01 139
1 (0.45) (−0.12) (−0.08) (0.56) (−2.43) (1.08) (0.21) (0.28) (−0.21) (−1.04) (−1.44) (0.99)
7.33 −1.74 −5.23 2.47 −3.76 0.04 −0.01 138 3.12 1.40 −3.50 −6.35 −4.05c −0.05 0.02 138
2 (0.84) (−0.41) (−1.16) (0.22) (−0.99) (0.66) (0.58) (0.54) (−1.26) (−0.93) (−1.73) (−1.10)
4.84 −0.54 −5.93 −9.86 −5.02 −0.02 0.04 137 −0.37 2.76 −2.12 −8.83 −5.06a −0.03 0.05 137
3 (0.75) (−0.17) (−1.81) (−1.23) (−1.75) (−0.35) (−0.09) (1.32) (−0.96) (−1.64) (−2.62) (−0.77)
4.49 −1.48 −2.40 −5.33 6.23 −0.01 −0.01 136 0.69 2.56 −3.33c −0.55 −0.82 −0.04 0.02 136
4 (0.84) (−0.56) (−0.85) (−0.80) (1.17) (−0.25) (0.19) (1.42) (−1.71) (−0.12) (−0.23) (−1.38)
J. Risk Financial Manag. 2021, 14, 585 30 of 39

Table A2. Cont.

Panel A
Forecast
Horizon Gold (1986:01–2020:12) Platinum (1986:04–2020:12)

(Y10 - ∆SP ∆EX ∆IP ∆EPU (Y10 - ∆SP ∆EX ∆IP ∆EPU
h C
Y3M ) R2 N C
Y3M ) R2 N

3.73 0.54 1.61 −1.05 −2.52 0.01 −0.02 138 3.32 −0.56 −3.60 −8.25 −1.92 0.02 −0.02 139
1 (0.86) (0.26) (0.73) (−0.20) (−1.37) (0.45) (0.46) (−0.16) (−0.96) (−0.91) (−0.61) (0.4)
5.19c 0.43 −2.09 2.21 −1.25 −0.03 0.01 137 3.83 −0.31 −6.05b −5.14 0.50 −0.03 0.02 138
2 (1.76) (0.30) (−1.37) (0.60) (−0.97) (−1.14) (0.79) (−0.13) (−2.43) (−0.84) (0.24) (−0.87)
4.68c 0.33 −0.91 1.45 −1.05 −0.02 −0.01 136 0.63 1.01 −2.32 −7.80c −1.71 −0.04 0.02 137
3 (1.84) (0.27) (−0.71) (0.46) (−0.93) (−0.85) (0.17) (0.55) (−1.21) (−1.66) (−1.02) (−1.44)
5.25b 0.38 −1.91 3.31 −2.83 −0.02 0.01 135 1.19 0.56 −1.71 −0.17 4.16 −0.04c 0.02 136
4 (2.34) (0.34) (−1.61) (1.19) (−1.27) (−1.22) (0.38) (0.36) (−1.03) (−0.04) (1.34) (−1.8)
Forecast
Horizon Palladium (1986:04–2020:12) Zinc (1997:08–2020:12)

(Y10 - ∆SP ∆EX ∆IP ∆EPU (Y10 - ∆SP ∆EX ∆IP ∆EPU
h C
Y3M ) R2 N C
Y3M ) R2 N

19.89c −6.40 −12.07b −7.36 8.63c 0.03 0.03 139 −1.27 0.56 −6.60 −15.05 0.07 0.17b 0.01 93
1 (1.94) (−1.29) (−2.27) (−0.57) (1.93) (0.37) (−0.13) (0.11) (−1.3) (−1.07) (0.02) (2.03)
16.52b −4.10 −7.36c −10.51 4.02 −0.05 0.02 138 −2.35 2.75 −2.46 −11.65 −0.71 0.03 −0.03 92
2 (2.16) (−1.11) (−1.87) (−1.1) (1.21) (−0.92) (−0.3) (0.7) (−0.61) (−1.05) (−0.23) (0.45)
11.65c −1.77 −0.97 −5.21 2.92 −0.03 −0.02 137 −6.15 5.27 −2.60 −5.86 −2.95 0.02 −0.01 91
3 (1.78) (−0.56) (−0.29) (−0.64) (1) (−0.69) (−0.91) (1.59) (−0.77) (−0.63) (−1.07) (0.43)
9.71c −1.60 0.49 −1.63 13.19b −0.03 0.01 136 −6.83 6.04b −1.16 7.09 3.45 0.02 0.01 90
4 (1.72) (−0.57) (0.16) (−0.23) (2.35) (−0.72) (−1.14) (2.05) (−0.37) (0.86) (0.57) (0.38)
Forecast
Horizon Ethanol (2005:06–2020:12) Coal (2009:01–2020:12)

(Y10 - ∆SP ∆EX ∆IP ∆EPU (Y10 - ∆SP ∆EX ∆IP ∆EPU
h C
Y3M ) R2 N C
Y3M ) R2 N

−1.39 −0.65 3.71 47.44c −7.37 0.20 0.07 62 4.07 −0.18 −3.93 −14.93 9.82b −0.12 0.10 48
1 (−0.07) (−0.07) (0.35) (1.95) (−1.12) (1.35) (0.26) (−0.03) (−0.54) (−0.84) (2.45) (−1.11)
−5.25 −0.63 1.61 −2.20 −7.50c 0.13 0.02 61 −1.02 2.98 −5.38 −16.73 5.05c −0.20b 0.15 47
2 (−0.43) (−0.1) (0.25) (−0.15) (−1.85) (1.47) (−0.09) (0.55) (−1.02) (−1.29) (1.72) (−2.42)
−5.53 0.68 −0.35 −8.83 −5.12c 0.01 −0.03 60 −13.96 7.81 −1.61 −12.06 −1.60 −0.13c 0.07 46
3 (−0.63) (0.16) (−0.08) (−0.83) (−1.7) (0.12) (−1.34) (1.66) (−0.36) (−1.1) (−0.61) (−1.84)
−3.89 0.66 −3.32 2.19 0.06 −0.02 −0.07 59 −12.55 6.48 −1.08 −7.51 1.56 −0.06 −0.01 45
4 (−0.58) (0.2) (−0.87) (0.27) (0.01) (−0.33) (−1.3) (1.48) (−0.24) (−0.74) (0.23) (−0.93)
Forecast
Horizon Natural gas (1990:05–2020:12)

(Y10 - ∆SP ∆EX ∆IP ∆EPU

h C
Y3M ) R2 N

5.82 −3.09 15.11c 15.35 1.91 0.00 0.00 122

1 (0.36) (−0.4) (1.87) (0.74) (0.29) (−0.01)
−0.62 0.29 2.48 −14.15 −3.37 0.02 −0.03 121
2 (−0.06) (0.05) (0.45) (−1) (−0.73) (0.22)
−6.55 2.69 −0.25 −26.80b −5.08 0.06 0.02 120
3 (−0.75) (0.65) (−0.06) (−2.46) (−1.38) (0.85)
−5.93 1.55 1.67 −16.60c 12.51c 0.04 0.03 119
4 (−0.86) (0.47) (0.47) (−1.93) (1.85) (0.7)
Panel B
Forecast
Horizon Oil (2004:01–2020:12) Silver (2004:01–2020:12)

(Y10 - ∆SP ∆EX ∆IP ∆EPU (Y10 - ∆SP ∆EX ∆IP ∆EPU
h C
Y3M ) R2 N C
Y3M ) R2 N

−0.10 −0.53 0.86 16.77 −15.95b 0.17 0.04 68 11.98 −4.06 −2.96 −13.37 −4.65 0.11 −0.03 68
1 (0.00) (−0.05) (0.07) (0.61) (−2.11) (1.01) (0.86) (−0.6) (−0.4) (−0.77) (−0.98) (1.09)
3.69 −2.15 0.72 9.54 −5.95 0.08 −0.03 67 16.38c −4.37 −7.96 −11.18 −3.03 −0.06 0.06 67
2 (0.25) (−0.3) (0.09) (0.52) (−1.2) (0.73) (1.77) (−0.98) (−1.63) (−0.98) (−0.97) (−0.9)
−0.32 0.45 −6.34 −16.19 −6.44c −0.02 0.03 66 9.86 −1.47 −4.98 −11.62 −4.22 −0.03 0.05 66
3 (−0.03) (0.08) (−1.1) (−1.21) (−1.68) (−0.27) (1.29) (−0.4) (−1.26) (−1.27) (−1.6) (−0.54)
−0.25 −0.17 −1.22 −1.81 8.51 −0.03 −0.05 65 10.70c −0.62 −6.94c 3.41 −0.20 −0.06 0.03 65
4 (−0.03) (−0.04) (−0.24) (−0.17) (1.04) (−0.44) (1.7) (−0.21) (−1.98) (0.45) (−0.03) (−1.23)
Forecast
Horizon Gold (2004:01–2020:12) Platinum (2004:01–2020:12)

(Y10 - ∆SP ∆EX ∆IP ∆EPU (Y10 - ∆SP ∆EX ∆IP ∆EPU
h C
Y3M ) R2 N C
Y3M ) R2 N

13.40b −2.83 −0.20 3.97 −1.68 0.02 −0.03 68 7.99 −5.32 −5.15 0.76 −1.60 0.13 −0.02 68
1 (2.15) (−0.94) (−0.06) (0.51) (−0.79) (0.45) (0.66) (−0.9) (−0.79) (0.05) (−0.39) (1.41)
15.01a −2.91 −3.22 0.05 −0.50 −0.03 0.02 67 11.86 −5.76 −10.43b −1.19 1.81 −0.01 0.05 67
2 (3.24) (−1.31) (−1.31) (0.01) (−0.32) (−0.96) (1.44) (−1.45) (−2.4) (−0.12) (0.65) (−0.17)
14.24a −2.59 −1.62 1.16 −0.16 −0.02 −0.02 66 7.39 −3.57 −4.49 −9.43 −1.74 −0.05 0.05 66
3 (3.44) (−1.31) (−0.76) (0.23) (−0.11) (−0.67) (1.16) (−1.17) (−1.36) (−1.23) (−0.79) (−1.04)
15.03a −2.38 −2.94 5.51 −1.90 −0.03 0.06 65 7.35 −3.13 −3.70 2.13 2.23 −0.05 0.02 65
4 (4.32) (−1.42) (−1.52) (1.31) (−0.6) (−1.29) (1.43) (−1.26) (−1.29) (0.34) (0.47) (−1.31)
J. Risk Financial Manag. 2021, 14, 585 31 of 39

Table A2. Cont.

Panel B
Forecast
Horizon Palladium (2004:01–2020:12) Zinc (2004:01–2020:12)

(Y10 - ∆SP ∆EX ∆IP ∆EPU (Y10 - ∆SP ∆EX ∆IP ∆EPU
h C
Y3M ) R2 N C
Y3M ) R2 N

25.75 −8.22 −8.27 −5.39 6.03 0.15 −0.01 68 8.68 −4.06 −15.79b −15.38 2.54 0.22b 0.06 68
1 (1.66) (−1.09) (−0.99) (−0.28) (1.14) (1.26) (0.66) (−0.63) (−2.23) (−0.93) (0.56) (2.19)
29.00a −8.33 −8.79 −15.25 2.57 −0.06 0.03 67 5.83 −0.67 −6.85 −9.64 0.78 0.03 −0.05 67
2 (2.64) (−1.58) (−1.51) (−1.13) (0.69) (−0.73) (0.53) (−0.13) (−1.19) (−0.72) (0.21) (0.37)
23.67b −5.24 −2.35 −12.46 0.92 −0.07 −0.01 66 −0.61 3.31 −5.75 −3.17 −1.63 0.01 −0.04 66
3 (2.54) (−1.18) (−0.49) (−1.11) (0.29) (−1.06) (−0.06) (0.73) (−1.17) (−0.28) (−0.5) (0.13)
21.04a −3.07 −3.58 −4.10 1.08 −0.07 −0.02 65 −2.05 4.53 −2.82 10.35 3.08 0.00 −0.03 65
4 (2.65) (−0.8) (−0.81) (−0.43) (0.15) (−1.26) (−0.24) (1.12) (−0.6) (1.02) (0.4) (0)
Forecast
Horizon Natural gas (2004:01–2020:12)

(Y10 - ∆SP ∆EX ∆IP ∆EPU

h C
Y3M ) R2 N

4.42 −5.06 17.53c 29.15 0.44 −0.03 −0.01 68

1 (0.23) (−0.54) (1.68) (1.2) (0.07) (−0.2)
1.56 −5.31 2.90 −6.76 −4.44 0.06 −0.05 67
2 (0.11) (−0.77) (0.38) (−0.38) (−0.92) (0.58)
−1.68 −4.13 −1.47 −18.22 −5.65 0.04 −0.01 66
3 (−0.15) (−0.75) (−0.25) (−1.32) (−1.43) (0.51)
−0.06 −5.29 −0.68 −10.39 11.01 0.03 −0.01 65
4 (−0.01) (−1.23) (−0.14) (−0.96) (1.35) (0.42)

Notes: The forecast horizon (h) is in quarters. Y10 -Y3M denotes the yield spread calculated as the difference between the yield rates on
10-year and 3-month government bonds. The table reports the estimation results of Equation (1) with the Newey and West (1987) procedure.
The sample period appears separately for each commodity. Figures in parentheses denote estimated standard errors. a, b and c denote
statistical significance at the 1%, 5%, and 10% levels, respectively.

Table A3. Estimation results of Equation (1) with the Y10 -Y1 indicator. Panel A: sample period 1986–2003. Panel B: sample
period 2004–2020.
Panel A
Forecast
Horizon Oil (1986:01–2003:12) Silver (1986:01–2003:12)

h C (Y10 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y10 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N
10.33 **
1 10.68 −5.04 0.38 −2.13 10.18 0.04 −0.06 72 −12.36 (0.03)
0.09 −11.41 −3.41 0.06 0.02 72
2 9.29 −2.61 −9.56 * −5.23 9.22 0.00 0.00 72
−10.04 * 8.40 ***
0.54 −6.30 −1.82 −0.01 0.05 72
(0.07) (0.05) (0.01)
3 7.75 −0.90 −5.37 −4.32 1.16 0.03 −0.03 72 −10.75 *** 8.55 *** 0.40 −9.64 * −2.06 0.00 0.13 72
(0.01) (0.00) (0.09)
4 6.47 −0.83 −3.00 −7.62 1.38 0.04 −0.03 72
−10.23 *** 8.30 ***
−0.13 −7.98 * −1.94 0.00 0.18 72
(0.00) (0.00) (0.09)
Forecast
Horizon Gold (1986:01–2003:12) Platinum (1986:04–2003:12)

h C (Y10 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y10 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N
1 −8.47 5.17 4.41 −10.64 3.55 0.01 0.03 70 −3.87 4.24 −1.97 −22.85 7.19 −0.07 0.05 72
2 −4.65 5.43 *** −1.55 2.89 −4.56 −0.01 0.07 70 −4.31 4.94 * −1.66 −14.42 ** 3.87 −0.03 0.07 72
(0.01) (0.08) (0.03)
3
−4.99 * 4.71 ***
−0.59 0.26 −2.62 −0.01 0.09 70 −5.63 5.03 **
0.32
−10.42 * 6.87 −0.01 0.11 72
(0.05) (0.00) (0.02) (0.05)
4
−5.03 ** 4.71 ***
−0.98 −0.86 −3.38 −0.01 0.12 70 −5.16 5.25 ***
−0.22 −5.71 5.58 −0.02 0.11 72
(0.03) (0.00) (0.01)
Forecast
Horizon Palladium (1986:04–2003:12) Zinc (1997:08–2003:12)

h C (Y10 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y10 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 11.59 −7.16 −14.61 * −12.90 33.00 **

−0.09 0.08 72
−21.33 * 9.16 8.23 −25.35 11.31 0.03 0.10 26
(0.05) (0.04) (0.06)
2 7.05 −5.81 −3.70 −11.63 31.04 ***
−0.07 0.06 72
−16.97 ** 7.48 6.01 −27.80 6.11 0.01 0.16 26
(0.01) (0.04)
3 3.32 −4.93 3.38 −4.13 36.35 ***
−0.02 0.14 72
−15.35 ** 7.55 **
3.84
−25.00 * −0.50 0.08 0.23 26
(0.00) (0.02) (0.04) (0.09)
4 1.79 −3.31 3.58 −4.30 30.21 *** −0.02 0.11 72 −15.03 *** 8.66 *** 2.34 −12.80 3.54 0.09 0.28 26
(0.00) (0.01) (0.01)
Forecast
Horizon Natural gas (1990:05–2003:12)

h C (Y10 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 4.59 −4.38 18.18 −12.43 36.19 0.01 −0.04 55
2 −3.53 3.48 5.06 −33.81 22.93 −0.03 −0.03 55
3 −12.73 7.90 5.30
−49.78 *** 23.53 *
0.11 0.15 55
(0.00) (0.09)
4 −9.98 8.53 4.29
−31.91 ** 14.60 0.09 0.10 55
(0.02)
J. Risk Financial Manag. 2021, 14, 585 32 of 39

Table A3. Cont.

Panel B
Forecast
Horizon Oil (2004:01–2020:12) Silver (2004:01–2020:12)

h C (Y10 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y10 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 0.01 −0.71 0.87 16.72

−15.96 ** 0.17 0.04 67 13.07 −6.08 −2.52 −14.07 −4.75 0.08 −0.03 67
(0.04)
2 3.06 −2.22 0.81 9.33
−6.02 * 0.07 −0.03 66
15.64 *
−4.56 −7.99 −11.44 −3.10 −0.06 0.06 66
(0.09) (0.09)
3 2.19 −1.56 −6.05 −16.46 −6.38 −0.04 0.04 65 10.99 −2.58 −4.89 −11.82 −4.20 −0.04 0.06 65
4 1.64 −1.67 −1.05 −1.99 8.65 −0.04 −0.05 64
11.94 *
−1.64 −6.86 * 3.29 −0.10 −0.06 0.03 64
(0.05) (0.06)
Forecast
Horizon Gold (2004:01–2020:12) Platinum (2004:01–2020:12)

h C (Y10 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y10 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N
12.91 **
1
(0.03)
−2.92 −0.23 3.82 −1.73 0.02 −0.03 67 6.12 −5.08 −5.07 0.34 −1.76 0.12 −0.03 67
2
14.64 ***
−3.10 −3.25 −0.10 −0.54 −0.03 0.02 66 9.76 −5.21 −10.51 ** −1.47 1.68 −0.01 0.04 66
(0.00) (0.02)
3 14.14 *** −2.91 −1.65 1.01 −0.19 −0.02 −0.02 65 7.63 −4.39 −4.44 −9.72 −1.78 −0.05 0.06 65
(0.00)
15.12 ***
4 (0.00) −2.78 −2.98 5.36 −2.01 −0.03 0.07 64 7.83 −4.00 −3.69 1.85 2.06 −0.05 0.03 64

Forecast
Horizon Palladium (2004:01–2020:12) Zinc (2004:01–2020:12)

h C (Y10 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y10 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1
24.57 *
−9.53 −7.76 −6.41 5.78 0.10 −0.02 67 9.45 −5.33 −15.72 ** −15.73 2.51
0.21 **
0.06 67
(0.09) (0.03) (0.04)
26.18 **
2
(0.02)
−7.80 −8.83 −15.73 2.36 −0.06 0.02 66 6.95 −1.46 −6.89 −9.66 0.83 0.03 −0.05 66
23.33 ***
3
(0.01)
−5.92 −2.31 −12.83 0.82 −0.07 −0.01 65 2.55 1.68 −5.68 −3.06 −1.41 0.01 −0.05 65
4 21.35 *** −3.68 −3.65 −4.26 1.01 −0.07 −0.02 64 1.63 2.63 −2.57 10.56 3.89 −0.01 −0.05 64
(0.00)
Forecast
Horizon Natural Gas (2004:01–2020:12)

h C (Y10 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 1.86 −3.96 17.37 28.99 0.32 −0.02 −0.01 67
2 −0.63 −4.66 2.84 −7.03 −4.58 0.06 −0.05 66
3 −2.01 −4.66 −1.41 −18.54 −5.74 0.04 −0.01 65
4 −1.42 −5.19 −0.79 −10.76 10.45 0.03 −0.01 64

Notes: The forecast horizon (h) is in quarters. Y10 -Y1 denotes the yield spread calculated as the difference between the yield rates on
10-year and 1-year government bonds. The table reports the estimation results of Equation (1) with the Newey and West (1987) procedure.
The sample period appears separately for each commodity. Figures in parentheses denote estimated standard errors. ***, ** and * denote
statistical significance at the 1%, 5%, and 10% levels, respectively.

Table A4. Estimation results of Equation (1) with the Y10 -Y2 indicator. Panel A: sample period 1986–2003. Panel B: sample
period 2004–2020.
Panel A
Forecast
Horizon Oil (1986:01–2003:12) Silver (1986:01–2003:12)

h C (Y10 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y10 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N
13.14 **
1 10.51 −6.94 0.47 −2.11 9.52 0.04 −0.06 72 −11.02 (0.02)
−0.01 −11.22 −1.76 0.06 0.03 72
2 8.54 −2.87 −9.57 * −5.37 8.69 0.00 0.00 72
−9.15 * 10.90 ***
0.44 −6.19 −0.53 −0.01 0.07 72
(0.07) (0.05) (0.00)
3 6.92 −0.37 −5.42 −4.51 0.81 0.03 −0.03 72 −10.21 *** 11.50 *** 0.26
−9.61 * −0.86 0.00 0.17 72
(0.01) (0.00) (0.08)
4 5.07 0.34 −3.11 −7.94 0.88 0.04 −0.03 72 −9.70*** 11.15 *** −0.26 −7.95 * −0.77 0.00 0.23 72
(0.00) (0.00) (0.08)
Forecast
Horizon Gold (1986:01–2003:12) Platinum (1986:04–2003:12)

h C (Y10 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y10 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 −8.04 6.88 *
4.32 −10.59 4.28 0.01 0.04 70 −3.70 5.81 −2.05 −22.86 ** 7.76 −0.07 0.05 72
(0.08) (0.03)
2 −4.12 7.13 ***
−1.63 2.96 −3.76 −0.01 0.09 70 −3.63 6.24 *
−1.71 −14.32 ** 4.67 −0.03 0.07 72
(0.00) (0.07) (0.03)
3
−4.85 ** 6.54 ***
−0.69 0.25 −2.02 −0.01 0.13 70 −5.00 6.41 **
0.27
−10.32 * 7.67 *
−0.01 0.12 72
(0.03) (0.00) (0.02) (0.05) (0.07)
4
−4.94 *** 6.59 ***
−1.08 −0.89 −2.80 −0.01 0.18 70 −4.48 6.69 ***
−0.27 −5.61 6.42 *
−0.02 0.12 72
(0.01) (0.00) (0.01) (0.09)
Forecast
Horizon Palladium (1986:04–2003:12) Zinc (1997:08–2003:12)

h C (Y10 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y10 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 9.00 −7.31 −14.69 −13.43 31.39 −0.09 0.07 72 −19.80 * 10.45 8.30 −24.92 12.87 0.03 0.10 26
(0.06)
2 6.24 −7.35 −3.65 −11.75 30.10 −0.07 0.06 72 −16.40 ** 9.25 * 5.98 −27.08 7.58 0.00 0.17 26
(0.04) (0.08)
3 2.72 −6.32 3.43 −4.21 35.57 −0.02 0.14 72
−14.85 ** 9.42 **
3.79
−24.23 * 1.01 0.08 0.26 26
(0.02) (0.03) (0.09)
4 1.38 −4.23 3.61 −4.36 29.68 −0.02 0.11 72
−14.53 *** 10.91 ***
2.28 −11.87 5.29 0.08 0.33 26
(0.01) (0.00)
J. Risk Financial Manag. 2021, 14, 585 33 of 39

Table A4. Cont.

Panel A
Forecast
Horizon Natural Gas (1990:05–2003:12)

h C (Y10 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 4.28 −5.76 18.25 −12.59 35.60 0.01 −0.04 55
2 −0.05 1.51 5.22 −32.94 23.99 −0.04 −0.03 55
3 −8.32 6.74 5.43
−48.61 *** 25.29 *
0.10 0.13 55
(0.01) (0.07)
4 −6.13 8.14 4.38 −30.86 ** 16.34 0.08 0.08 55
(0.03)
Panel B
Forecast
Horizon Oil (2004:01–2020:12) Silver (2004:01–2020:12)

h C (Y10 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y10 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 0.64 −1.32 0.90 16.67

−15.94 ** 0.17 0.04 67 12.44 −6.54 −2.47 −14.23 −4.82 0.08 −0.03 67
(0.04)
2 2.39 −2.04 0.81 9.29 −6.07 0.07 −0.03 66 15.43 * −5.12 −7.94 −11.58 −3.15 −0.06 0.05 66
(0.09)
3 2.44 −2.01 −6.02 −16.53 −6.38 * −0.04 0.04 65 12.12 −3.87 −4.83 −11.96 −4.17 −0.04 0.06 65
(0.09)
4 2.98 −2.99 −0.97 −2.09 8.80 −0.04 −0.05 64
13.44 **
−3.08 −6.78 * 3.18 0.08 −0.06 0.04 64
(0.03) (0.06)
Forecast
Horizon Gold (2004:01–2020:12) Platinum (2004:01–2020:12)

h C (Y10 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y10 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N
12.82 **
1
(0.04)
−3.31 −0.20 3.73 −1.76 0.02 −0.03 67 4.78 −4.82 −5.04 0.25 −1.84 0.12 −0.03 67
2
14.76 ***
−3.68 −3.21 −0.21 −0.56 −0.03 0.02 66 8.53 −5.06 −10.48 ** −1.57 1.59 −0.01 0.04 66
(0.00) (0.02)
14.61 ***
3
(0.00)
−3.74 −1.60 0.89 −0.20 −0.02 −0.01 65 7.96 −5.35 −4.37 −9.88 −1.82 −0.05 0.06 65
4 15.67 *** −3.66 * −2.93 5.24 −2.01 −0.03 0.08 64 8.58 * −5.23 −3.62 1.69 2.05 −0.05 0.04 64
(0.00) (0.09) (0.09)
Forecast
Horizon Palladium (2004:01–2020:12) Zinc (2004:01–2020:12)

h C (Y10 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y10 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 23.20 −9.94 −7.69 −6.65 5.65 0.10 −0.02 67 10.90 −7.32 −15.61 ** −15.98 2.51
0.21 **
0.06 67
(0.03) (0.04)
25.65 **
2
(0.02)
−8.61 −8.75 −15.96 2.27 −0.06 0.02 66 9.24 −3.49 −6.80 −9.83 0.89 0.03 −0.05 66
24.84 ***
3
(0.01)
−8.04 −2.20 −13.10 0.83 −0.08 0.00 65 5.34 −0.22 −5.64 −3.14 −1.25 0.01 −0.05 65
23.42 ***
4
(0.00)
−5.90 −3.52 −4.47 1.21 −0.07 −0.01 64 4.68 0.63 −2.45 10.55 4.42 −0.01 −0.06 64

Forecast
Horizon Natural Gas (2004:01–2020:12)

h C (Y10 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 2.17 −4.83 17.42 28.85 0.30 −0.02 −0.01 67
2 −0.89 −5.19 2.89 −7.17 −4.63 0.06 −0.05 66
3 −1.55 −5.77 −1.34 −18.72 −5.77 0.04 −0.01 65
4 −1.25 −6.14 −0.74 −10.95 10.32 0.03 −0.01 64

Notes: The forecast horizon (h) is in quarters. Y10 -Y2 denotes the yield spread calculated as the difference between the yield rates on
10-year and 2-year government bonds. The table reports the estimation results of Equation (1) with the Newey and West (1987) procedure.
The sample period appears separately for each commodity. Figures in parentheses denote estimated standard errors. ***, ** and * denote
statistical significance at the 1%, 5%, and 10% levels, respectively.

Table A5. Estimation results of Equation (1) with the Y30 -Y3M indicator. Panel A: sample period 1986–2003. Panel B: sample
period 2004–2020.
Panel A
Forecast
Horizon Oil (1986:01–2003:12) Silver (1986:01–2003:12)

(Y30 - ∆SP ∆EX ∆IP ∆EPU (Y30 - ∆SP ∆EX ∆IP ∆EPU
h C
M3 ) R2 N C
M3 ) R2 N

7.41 **
1 12.77 −4.45 0.56 −1.91 11.52 0.04 −0.06 72 −13.32 (0.03)
−0.04 −11.21 −4.69 0.05 0.01 72
2 9.63 −1.93 −9.52 * −5.25 9.60 0.00 0.00 72 −11.32 ** 6.28 *** 0.39 −6.24 −3.08 −0.02 0.05 72
(0.07) (0.04) (0.01)
3 8.10 −0.78 −5.34 −4.29 1.39 0.03 −0.03 72
−12.25 *** 6.50 ***
0.24
−9.61 * −3.43 −0.01 0.14 72
(0.01) (0.00) (0.09)
4 6.17 −0.41 −3.02 −7.71 1.32 0.04 −0.03 72
−11.34 *** 6.13 ***
−0.26 −7.89 * −3.12 −0.01 0.17 72
(0.00) (0.00) (0.09)
Forecast
Horizon Gold (1986:01–2003:12) Platinum (1986:04–2003:12)

(Y30 - ∆SP ∆EX ∆IP ∆EPU (Y30 - ∆SP ∆EX ∆IP ∆EPU
h C
M3 ) R2 N C
M3 ) R2 N

1 −7.98 3.22 4.39 −10.34 3.31 0.01 0.02 70 −5.10 3.47 −2.09 −22.94 6.29 −0.07 0.05 72
2 −4.92 3.77 ***
−1.60 3.03 −5.13 −0.02 0.06 70 −5.62 3.98 *
−1.79 −14.49 ** 2.88 −0.04 0.08 72
(0.01) (0.06) (0.03)
3
−5.12 * 3.22 ***
−0.63 0.41 −3.06 −0.02 0.07 70
−6.89 * 4.02 ***
0.20
−10.48 ** 5.88 −0.01 0.12 72
(0.06) (0.01) (0.08) (0.01) (0.04)
4 −4.96 ** 3.11 *** −1.00 −0.68 −3.73 −0.01 0.09 70 −6.42 * 4.17 *** −0.34 −5.76 4.59 −0.02 0.12 72
(0.04) (0.00) (0.07) (0.01)
J. Risk Financial Manag. 2021, 14, 585 34 of 39

Table A5. Cont.

Panel A
Forecast
Horizon Palladium (1986:04–2003:12) Zinc (1997:08–2003:12)

(Y30 - ∆SP ∆EX ∆IP ∆EPU (Y30 - ∆SP ∆EX ∆IP ∆EPU
h C
M3 ) R2 N C
M3 ) R2 N

1 10.62 −4.30 −14.65 * −13.36 33.17 **

−0.08 0.07 72
−21.63 * 6.28 7.90 −25.51 9.97 0.03 0.09 26
(0.05) (0.04) (0.07)
2 7.04 −3.89 −3.67 −11.85 31.52 ***
−0.07 0.06 72
−18.24 ** 5.68 *
5.59 −27.52 5.02 0.01 0.17 26
(0.01) (0.04) (0.09)
3 3.00 −3.14 3.38 −4.38 36.62 ***
−0.01 0.13 72
−17.30 *** 6.09 **
3.31
−24.44 * −1.61 0.08 0.27 26
(0.00) (0.01) (0.02) (0.09)
4 0.66 −1.64 3.51 −4.65 29.98 ***
−0.02 0.11 72
−17.16 *** 6.94 ***
1.75 −12.20 2.27 0.09 0.34 26
(0.00) (0.01) (0.00)
Forecast
Horizon Natural Gas (1990:05–2003:12)

(Y30 - ∆SP ∆EX ∆IP ∆EPU

h C
M3 ) R2 N

1 7.71 −4.40 18.39 −11.95 37.52 0.01 −0.04 55

2 −3.67 2.41 5.02 −33.69 22.62 −0.03 −0.03 55
3 −14.50 6.16 5.13
−49.82 *** 22.36 0.10 0.15 55
(0.00)
4 −13.07 7.20 *
4.04
−32.21 ** 12.97 0.09 0.11 55
(0.07) (0.02)
Panel B
Forecast
Horizon Oil (2004:01–2020:12) Silver (2004:01–2020:12)

(Y30 - ∆SP ∆EX ∆IP ∆EPU (Y30 - ∆SP ∆EX ∆IP ∆EPU
h C
M3 ) R2 N C
M3 ) R2 N

1 −3.50 1.08 0.76 16.84

−16.06 ** 0.17 0.04 67 15.85 −5.05 −2.31 −14.20 −4.77 0.08 −0.03 67
(0.04)
17.66 *
2 2.08 −0.98 0.81 9.35 −6.09 0.07 −0.03 66
(0.08)
−3.76 −7.83 −11.53 −3.12 −0.06 0.06 66
3 0.66 −0.34 −6.05 −16.43 −6.48 * −0.04 0.04 65 13.20 −2.58 −4.77 −11.91 −4.17 −0.04 0.06 65
(0.09)
4 0.65 −0.63 −1.06 −1.98 8.48 −0.04 −0.05 64
13.85 *
−1.87 −6.70 * 3.26 0.23 −0.06 0.04 64
(0.05) (0.06)
Forecast
Horizon Gold (2004:01–2020:12) Platinum (2004:01–2020:12)

(Y30 - ∆SP ∆EX ∆IP ∆EPU (Y30 - ∆SP ∆EX ∆IP ∆EPU
h C
M3 ) R2 N C
M3 ) R2 N

15.19 **
1
(0.03)
−2.83 −0.11 3.73 −1.71 0.02 −0.03 67 9.44 −4.65 −4.86 0.20 −1.75 0.12 −0.02 67
2
16.84 ***
−2.91 −3.12 −0.19 −0.52 −0.03 0.04 66 12.79 −4.61 −10.31 ** −1.60 1.68 −0.01 0.05 66
(0.00) (0.02)
3
16.44 *** −2.83 * −1.52 0.92 −0.16 −0.02 0.01 65 9.74 −3.69 −4.28 −9.82 −1.80 −0.05 0.07 65
(0.00) (0.08)
4
17.09 *** −2.63 * −2.79 5.32 −1.67 −0.03 0.09 64
9.50 *
−3.28 −3.48 1.81 2.35 −0.05 0.04 64
(0.00) (0.05) (0.09)
Forecast
Horizon Palladium (2004:01–2020:12) Zinc (2004:01–2020:12)

(Y30 - ∆SP ∆EX ∆IP ∆EPU (Y30 - ∆SP ∆EX ∆IP ∆EPU
h C
M3 ) R2 N C
M3 ) R2 N

1
29.84 *
−8.31 −7.40 −6.65 5.77 0.10 −0.01 67 8.58 −2.99 −15.64 ** −15.73 2.41 0.21 0.05 67
(0.07) (0.03)
30.47 ***
2
(0.01)
−6.79 −8.53 −15.93 2.35 −0.06 0.03 66 5.39 −0.25 −6.91 −9.61 0.76 0.03 −0.05 66
26.31 ***
3
(0.01)
−5.03 −2.10 −12.97 0.81 −0.07 0.00 65 −0.13 2.21 −5.79 −2.97 −1.49 0.01 −0.04 65
4 22.79 *** −2.97 −3.46 −4.30 1.26 −0.07 −0.02 64 −1.41 3.00 −2.82 10.61 3.38 −0.01 −0.04 64
(0.01)
Forecast
Horizon Natural gas (2004:01–2020:12))

(Y30 - ∆SP ∆EX ∆IP ∆EPU

h C
M3 ) R2 N

1 0.34 −1.84 17.39 29.02 0.22 −0.02 −0.01 67

2 −1.16 −2.72 2.91 −7.04 −4.67 0.06 −0.06 66
3 −2.99 −2.55 −1.33 −18.55 −5.89 0.04 −0.01 65
4 −1.73 −3.17 −0.66 −10.78 10.40 0.03 −0.02 64

Notes: The forecast horizon (h) is in quarters. Y30 -Y3M denotes the yield spread calculated as the difference between the yield rates on
10-year and 3-month government bonds. The table reports the estimation results of Equation (1) with the Newey and West (1987) procedure.
The sample period appears separately for each commodity. Figures in parentheses denote estimated standard errors. ***, ** and * denote
statistical significance at the 1%, 5%, and 10% levels, respectively.
J. Risk Financial Manag. 2021, 14, 585 35 of 39

Table A6. Estimation results of Equation (1) with the Y30 -Y1 indicator. Panel A: sample period 1986–2003. Panel B: sample
period 2004–2020.
Panel A
Forecast
Horizon Oil (1986:01–2003:12) Silver (1986:01–2003:12)

h C (Y30 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y30 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 11.73 −4.60 0.50 −1.93 10.11 0.04 −0.06 72

−13.51 * 8.82 ***
−0.08 −11.60 −2.95 0.06 0.03 72
(0.08) (0.01)
2 9.08 −1.93 −9.55 * −5.29 8.95 0.00 0.00 72
−11.34 ** 7.39 ***
0.37 −6.54 −1.56 −0.01 0.08 72
(0.07) (0.03) (0.00)
3 7.17 −0.36 −5.41 −4.46 0.90 0.03 −0.03 72
−12.32 *** 7.68 ***
0.21
−9.93 * −1.88 0.00 0.19 72
(0.00) (0.00) (0.07)
4 5.40 −0.01 −3.08 −7.86 0.97 0.04 −0.04 72 −11.52 *** 7.31 *** −0.30 −8.22 * −1.69 0.00 0.24 72
(0.00) (0.00) (0.07)
Forecast
Horizon Gold (1986:01–2003:12) Platinum (1986:04–2003:12)

h C (Y30 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y30 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 −8.48 4.09 4.35 −10.60 3.93 0.01 0.03 70 −4.59 3.77 −2.06 −22.99 ** 7.30 −0.07 0.06 72
(0.03)
2 −4.99 4.49 ***
−1.62 2.85 −4.25 −0.01 0.09 70 −4.77 4.17 *
−1.74 −14.49 ** 4.12 −0.03 0.08 72
(0.00) (0.06) (0.03)
7.05
3
−5.71 ** 4.16 ***
−0.69 0.14 −2.48 −0.01 0.13 70
−6.35 * 4.39 ***
0.22
−10.54 ** * −0.01 0.13 72
(0.02) (0.00) (0.08) (0.01) (0.04) (0.09)
−5.69 4.11 *** −5.93 * 4.60 ***
4 *** (0.00) −1.07 −0.97 −3.23 −0.01 0.16 70 (0.07) (0.00) −0.32 −5.85 5.76 −0.02 0.13 72
(0.01)
Forecast
Horizon Palladium (1986:04–2003:12) Zinc (1997:08–2003:12)

h C (Y30 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y30 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 10.57 −5.02 −14.64 * −13.17 32.11 ** −0.09 0.08 72 −21.84 * 6.90 8.09 −25.65 11.71 0.03 0.10 26
(0.05) (0.04) (0.06)
2 7.38 −4.78 −3.63 −11.59 30.69 ***
−0.07 0.07 72
−18.30 ** 6.16 *
5.78 −27.69 6.57 0.00 0.18 26
(0.01) (0.03) (0.08)
3 3.17 −3.79 3.40 −4.19 35.91 ***
−0.02 0.14 72
−17.09 *** 6.44 **
3.55
−24.73 * 0.02 0.08 0.28 26
(0.00) (0.01) (0.02) (0.08)
4 1.10 −2.19 3.55 −4.48 29.72 ***
−0.02 0.11 72
−17.03 *** 7.41 ***
2.01 −12.49 4.14 0.08 0.35 26
(0.00) (0.00) (0.00)
Forecast
Horizon Natural gas (1990:05–2003:12)

h C (Y30 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 6.68 −4.48 18.32 −12.18 36.38 0.01 −0.04 55
2 −1.62 1.67 5.15 −33.25 23.58 −0.04 −0.03 55
3 −11.31 5.34 5.33
−49.12 *** 24.34 *
0.10 0.14 55
(0.01) (0.08)
4 −8.91 6.01 4.30 −31.30 ** 15.38 0.09 0.09 55
(0.03)
Panel B
Forecast
Horizon Oil (2004:01–2020:12) Silver (2004:01–2020:12)

h C (Y30 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y30 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 −3.42 1.15 0.77 16.91

−16.04 ** 0.17 0.04 67 14.08 −4.73 −2.40 −14.42 −4.86 0.08 −0.03 67
(0.04)
2 0.85 −0.50 0.77 9.36 −6.13 0.07 −0.03 66 16.73 * −3.71 −7.89 −11.72 −3.19 −0.06 0.06 66
(0.09)
3 1.44 −0.75 −6.04 −16.50 −6.46 * −0.04 0.04 65
13.60 *
−3.04 −4.77 −12.10 −4.19 −0.03 0.06 65
(0.09) (0.09)
4 1.53 −1.12 −1.02 −2.06 8.57 −0.04 −0.05 64
14.52 **
−2.39 −6.70 * 3.10 0.22 −0.06 0.04 64
(0.03) (0.06)
Forecast
Horizon Gold (2004:01–2020:12) Platinum (2004:01–2020:12)

h C (Y30 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y30 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N
14.67 **
1
(0.03)
−2.87 −0.14 3.57 −1.76 0.02 −0.03 67 6.63 −3.78 −4.97 0.07 −1.86 0.12 −0.03 67
2
16.43 ***
−3.02 −3.15 −0.36 −0.57 −0.03 0.04 66 10.72 −4.10 −10.40 ** −1.77 1.58 −0.01 0.04 66
(0.00) (0.02)
3
16.26 *** −3.04 * −1.54 0.74 −0.21 −0.02 0.01 65 9.37 −3.91 −4.31 −10.04 −1.87 −0.05 0.07 65
(0.00) (0.07)
4 17.07 *** −2.89 ** −2.83 5.14 −1.82 −0.03 0.10 64 9.44 * −3.58 * −3.53 1.58 2.15 −0.05 0.05 64
(0.00) (0.04) (0.08) (0.09)
Forecast
Horizon Palladium (2004:01–2020:12) Zinc (2004:01–2020:12)

h C (Y30 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y30 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 25.07 −6.88 −7.61 −6.89 5.57 0.11 −0.02 67 8.66 −3.34 −15.66 ** −15.93 2.38
0.21 **
0.05 67
(0.03) (0.04)
27.23 **
2
(0.02)
−5.95 −8.68 −16.16 2.20 −0.06 0.02 66 6.70 −0.90 −6.87 −9.71 0.79 0.03 −0.05 66
25.42 ***
3
(0.01)
−5.14 −2.15 −13.24 0.70 −0.07 0.00 65 2.66 1.13 −5.71 −2.99 −1.35 0.01 −0.05 65
23.11 ***
4
(0.00)
−3.43 −3.49 −4.53 1.13 −0.07 −0.02 64 1.69 1.82 −2.61 10.69 4.00 −0.01 −0.05 64
J. Risk Financial Manag. 2021, 14, 585 36 of 39

Table A6. Cont.

Panel B
Forecast
Horizon Natural gas (2004:01–2020:12))

h C (Y30 -Y1 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 −1.61 −1.10 17.31 29.03 0.16 −0.02 −0.01 67
2 −3.34 −1.97 2.82 −7.08 −4.75 0.06 −0.06 66
3 −3.84 −2.42 −1.37 −18.66 −5.96 0.04 −0.01 65
4 −3.23 −2.78 −0.79 −10.94 10.00 0.03 −0.02 64

Notes: The forecast horizon (h) is in quarters. Y30 -Y1 denotes the yield spread calculated as the difference between the yield rates on
10-year and 1-year government bonds. The table reports the estimation results of Equation (1) with the Newey and West (1987) procedure.
The sample period appears separately for each commodity. Figures in parentheses denote estimated standard errors. ***, ** and * denote
statistical significance at the 1%, 5%, and 10% levels, respectively.

Table A7. Estimation results of Equation (1) with the Y30 -Y2 indicator. Panel A: sample period 1986–2003. Panel B: sample
period 2004–2020.
Panel A
Forecast
Horizon Oil (1986:01–2003:12) Silver (1986:01–2003:12)

h C (Y30 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y30 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 11.32 −5.65 0.56 −1.99 9.36 0.04 −0.06 72

−11.97 * 10.26 ***
−0.13 −11.33 −1.33 0.06 0.04 72
(0.09) (0.01)
2 8.40 −1.97 −9.57 * −5.43 8.52 0.00 0.00 72
−10.23 ** 8.73 ***
0.31 −6.35 −0.24 −0.01 0.09 72
(0.07) (0.03) (0.00)
3 6.51 0.05 −5.46 −4.60 0.70 0.03 −0.03 72
−11.45 *** 9.29 ***
0.12
−9.80 * −0.58 0.00 0.22 72
(0.00) (0.00) (0.07)
4 4.35 0.81 −3.17 −8.10 0.73 0.04 −0.03 72
−10.68 *** 8.84 ***
−0.38 −8.09 * −0.44 0.00 0.28 72
(0.00) (0.00) (0.07)
Forecast
Horizon Gold (1986:01–2003:12) Platinum (1986:04–2003:12)

h C (Y30 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y30 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 −7.86 4.87 *
4.29 −10.48 4.62 0.01 0.03 70 −4.23 4.62 −2.11 −22.94 ** 7.92 −0.07 0.06 72
(0.09) (0.03)
2 −4.30 5.30 *** −1.66 2.97 −3.45 −0.01 0.10 70 −4.00 4.81 * −1.76 −14.35 ** 4.89 −0.04 0.08 72
(0.00) (0.05) (0.03)
−5.34 ** 5.13 *** −5.61 * 5.12 *** −10.41 ** 7.85
3
(0.02) (0.00)
−0.75 0.19 −1.82 −0.01 0.16 70
(0.09) (0.01)
0.19
(0.04)
* −0.01 0.13 72
(0.06)
−5.36 5.10 *** −5.15 * 5.36 ***
6.60
4 ***
(0.00)
−1.14 −0.92 −2.58 −0.01 0.21 70
(0.09) (0.00)
−0.35 −5.71 * −0.02 0.14 72
(0.00) (0.08)
Forecast
Horizon Palladium (1986:04–2003:12) Zinc (1997:08–2003:12)

h C (Y30 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y30 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N
−14.73 30.87 ** −20.44 *
1 8.32 −4.76 ** −13.64 (0.04)
−0.08 0.07 72
(0.07)
7.38 8.17 −25.51 12.79 0.03 0.09 26
(0.04)
2 6.49 −5.51 −3.60 −11.75 29.80 ***
−0.07 0.07 72
−17.71 ** 7.05 *
5.76 −27.27 7.72 0.00 0.19 26
(0.01) (0.03) (0.07)
3 2.48 −4.39 3.43 −4.32 35.21 ***
−0.02 0.14 72
−16.55 *** 7.43 **
3.52
−24.26 * 1.24 0.08 0.30 26
(0.00) (0.01) (0.02) (0.09)
4 0.64 −2.48 3.56 −4.56 29.30 ***
−0.02 0.11 72
−16.49 *** 8.59 ***
1.96 −11.91 5.56 0.08 0.39 26
(0.00) (0.00) (0.00)
Forecast
Horizon Natural Gas (1990:05–2003:12)

h C (Y30 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 6.22 −5.39 18.38 −12.43 35.71 0.01 −0.04 55
2 1.23 0.21 5.31 −32.64 24.24 −0.04 −0.03 55
3 −7.52 4.24 5.48
−48.20 *** 25.63 *
0.10 0.13 55
(0.01) (0.07)
4 −5.53 5.36 4.41
−30.44 ** 16.70 0.08 0.07 55
(0.03)
Panel B
Forecast
Horizon Oil (2004:01–2020:12) Silver (2004:01–2020:12)

h C (Y30 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y30 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 −3.48 1.30 0.75 16.95 −16.03 ** 0.17 0.04 67 13.47 −4.90 −2.36 −14.55 −4.93 0.08 −0.03 67
(0.04)
16.55 *
2 −0.15 −0.03 0.74 9.42 −6.15 0.07 −0.03 66 (0.09) −4.01 −7.84 −11.85 −3.23 −0.06 0.06 66
3 1.47 −0.84 −6.02 −16.53 −6.47 * −0.04 0.04 65
14.86 *
−4.02 −4.68 −12.32 −4.20 −0.03 0.07 65
(0.09) (0.07)
4 2.57 −1.79 −0.95 −2.18 8.67 −0.04 −0.05 64
16.11 **
−3.48 −6.58 * 2.90 0.34 −0.06 0.05 64
(0.02) (0.06)
J. Risk Financial Manag. 2021, 14, 585 37 of 39

Table A7. Cont.

Panel B
Forecast
Horizon Gold (2004:01–2020:12) Platinum (2004:01–2020:12)

h C (Y30 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y30 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N
14.78 **
1
(0.03)
−3.23 −0.09 3.45 −1.79 0.02 −0.03 67 5.38 −3.52 −4.97 0.03 −1.93 0.12 −0.03 67
2
16.70 ***
−3.48 −3.10 −0.50 −0.60 −0.03 0.04 66 9.64 −3.96 −10.39 ** −1.84 1.51 −0.01 0.04 66
(0.00) (0.02)
3 16.88 *** −3.69 * −1.48 0.58 −0.24 −0.02 0.02 65 9.76 −4.53 −4.24 −10.22 −1.92 −0.05 0.07 65
(0.00) (0.05)
4 17.75 *** −3.55 ** −2.76 4.97 −1.83 −0.03 0.11 64 10.18 * −4.35 * −3.45 1.39 2.12 −0.05 0.05 64
(0.00) (0.03) (0.06) (0.07)
Forecast
Horizon Palladium (2004:01–2020:12) Zinc (2004:01–2020:12)

h C (Y30 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N C (Y30 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 23.62 −6.84 −7.57 −7.03 5.47 0.11 −0.03 67 9.55 −4.16 −15.58 ** −16.12 2.35
0.21 **
0.05 67
(0.03) (0.04)
26.66 **
2
(0.03)
−6.28 −8.62 −16.34 2.11 −0.06 0.02 66 8.53 −1.96 −6.79 −9.89 0.81 0.04 −0.05 66
26.79 ***
3
(0.01)
−6.41 −2.03 −13.55 0.66 −0.07 0.01 65 5.05 0.00 −5.65 −3.13 −1.27 0.01 −0.05 65
24.98 ***
4
(0.00)
−4.78 −3.35 −4.79 1.26 −0.07 −0.01 64 4.33 0.61 −2.48 10.59 4.39 −0.01 −0.06 64

Forecast
Horizon Natural Gas (2004:01–2020:12)

h C (Y30 -Y2 ) ∆SP ∆EX ∆IP ∆EPU R2 N

1 −2.07 −0.97 17.31 29.02 0.14 −0.02 −0.01 67
2 −4.18 −1.73 2.81 −7.08 −4.79 0.06 −0.06 66
3 −3.91 −2.64 −1.33 −18.75 −6.00 0.04 −0.01 65
4 −3.57 −2.90 −0.78 −11.03 9.87 0.03 −0.03 64

Notes: The forecast horizon (h) is in quarters. Y30 -Y2 denotes the yield spread calculated as the difference between the yield rates on
10-year and 2-year government bonds. The table reports the estimation results of Equation (1) with the Newey and West (1987) procedure.
The sample period appears separately for each commodity. Figures in parentheses denote estimated standard errors. ***, ** and * denote
statistical significance at the 1%, 5%, and 10% levels, respectively.

Note
q
1 The conditional correlation between two random variables y1 and y2 is ρ12,t = Et−1 (y1t y2t )/ Et−1 (y21t ) Et−1 y22t . It is acceptable


√
to present returns as the conditional standard deviation times the standardized disturbance. yit = hit ε it . This is because
hit = Et−1 y2it . For each series i, ε it is a standardized disturbance with a mean of zero and a variance of one. Accordingly, the


q
conditional correlation can be presented as ρ12,t = Et−1 (ε 1t ε 2t )/ Et−1 (ε21t ) Et−1 ε22t = Et−1 (ε 1t ε 2t ). Hence, the conditional

correlation is also the conditional covariance between the standardized disturbances. This is the spirit of the DCC method.

References
Abdymomunov, Azamat. 2013. Predicting output using the entire yield curve. Journal of Macroeconomics 37: 333–44. [CrossRef]
Akram, Farooq. 2009. Commodity prices, interest rates and the dollar. Energy Economics 31: 838–51. [CrossRef]
Arango, Luis, Fernando Arias, and Adriana Flórez. 2011. Determinants of commodity prices. Applied Economics 44: 135–45. [CrossRef]
Bai, Jushan, and Pierre Perron. 2003. Critical values for multiple structural change tests. Econometrics Journal 6: 72–78. [CrossRef]
Bampinas, Georgios, and Theodore Panagiotidis. 2015. Are gold and silver a hedge against inflation? A two century perspective.
International Review of Financial Analysis 41: 267–76. [CrossRef]
Batten, Jonathan, Cetin Ciner, and Brian Lucey. 2010. The macroeconomic determinants of volatility in precious metals markets.
Resources Policy 35: 65–71. [CrossRef]
Beckmann, Joscha, and Robert Czudaj. 2013. Gold as an inflation hedge in a time-varying coefficient framework. North American
Journal of Economics and Finance 24: 208–22. [CrossRef]
Bodie, Zvi, and Victor Rosansky. 1980. Risk and return in commodity futures. Financial Analysts Journal 36: 27–39. [CrossRef]
Bollerslev, Tim. 1990. Modelling the coherence in short-run nominal exchange rates: A multivariate generalized ARCH model. Review
of Economics and Statistics 72: 498–505. [CrossRef]
Chevallier, Julien, Mathieu Gatumel, and Florian Ielpo. 2014. Commodity markets through the business cycle. Quantitative Finance 14:
1597–618. [CrossRef]
Chinn, Menzie, and Kavan Kucko. 2015. The predictive power of the yield curve across countries and time. International Finance 18:
129–56. [CrossRef]
Churchill, Sefa, John Inekwe, Kris Ivanovski, and Russell Smyth. 2019. Dynamics of oil price, precious metal prices and the exchange
rate in the long-run. Energy Economics 84: 104508. [CrossRef]
Dai, Zhifeng, and Jie Kang. 2021. Bond yield and crude oil prices predictability. Energy Economics 97: 105205. [CrossRef]
Daskalaki, Charoula, George Skiadopoulos, and Nikolas Topaloglou. 2017. Diversification benefits of commodities: A stochastic
dominance efficiency approach. Journal of Empirical Finance 44: 250–69. [CrossRef]
J. Risk Financial Manag. 2021, 14, 585 38 of 39

Demiralay, Sercan, Selcuk Bayraci, and Gaye Gencer. 2019. Time-varying diversification benefits of commodity futures. Empirical
Economics 56: 1823–53. [CrossRef]
Duarte, Angelo, Wagner Gaglianone, Osmani de Carvalho Guillén, and João Issler. 2021. Commodity Prices and Global Economic
Activity: A derived-demand approach. Energy Economics 96: 105120. [CrossRef]
Duffee, Gregory. 1998. The relation between treasury yields and corporate bond yield spreads. Journal of Finance 53: 2225–41. [CrossRef]
Engle, Robert. 2002. Dynamic conditional correlation: A simple class of multivariate generalized autoregressive conditional het-
eroskedasticity models. Journal of Business & Economic Statistics 20: 339–50.
Estrella, Arturo. 2005. Why does the yield curve predict output and inflation? Economic Journal 115: 722–44. [CrossRef]
Estrella, Arturo, and Gikas Hardouvelis. 1991. The term structure as a predictor of real economic activity. Journal of Finance 46: 555–76.
[CrossRef]
Evgenidis, Anastasios, Stephanos Papadamou, and Costas Siriopoulos. 2020. The yield spread’s ability to forecast economic activity:
What have we learned after 30 years of studies? Journal of Business Research 106: 221–32. [CrossRef]
Fama, Eugene, and Kenneth French. 1988. Business cycles and the behavior of metals prices. The Journal of Finance 43: 1075–93.
[CrossRef]
Fasanya, Ismail, and Crystal Awodimila. 2020. Are commodity prices good predictors of inflation? The African perspective. Resources
Policy 69: 101802. [CrossRef]
Gagnon, Marie, Guillaume Manseau, and Gabriel Power. 2020. They’re back! Post-financialization diversification benefits of
commodities. International Review of Financial Analysis 71: 101515. [CrossRef]
Garner, Alan. 1989. Commodity prices: Policy target or information variable?: Note. Journal of Money, Credit and Banking 21: 508–14.
[CrossRef]
Giacomini, Raffaella, and Barbara Rossi. 2006. How stable is the forecasting performance of the yield curve for output growth? Oxford
Bulletin of Economics and Statistics 68: 783–95. [CrossRef]
Gogas, Periklis, Theophilos Papadimitriou, Maria Matthaiou, and Efthymia Chrysanthidou. 2015. Yield curve and recession forecasting
in a machine learning framework. Computational Economics 45: 635–45.
Gorton, Gary, and Geert Rouwenhorst. 2006. Facts and fantasies about commodity futures. Financial Analysts Journal 62: 47–68.
[CrossRef]
Granger, Clive. 1969. Investigating causal relations by econometric models and cross-spectral methods. Econometrica: Journal of the
Econometric Society 37: 424–38. [CrossRef]
Hamilton, James, and Dong Kim. 2002. A reexamination of the predictability of economic activity using the yield spread. Journal of
Money, Credit, and Banking 34: 340–60. [CrossRef]
Hamilton, James, and Jing Wu. 2015. Effects of index-fund investing on commodity futures prices. International Economic Review 56:
187–205. [CrossRef]
Harvey, Campbell. 1989. Forecasts of economic growth from the bond and stock markets. Financial Analysts Journal 45: 38–45.
[CrossRef]
Harvey, David, Neil Kellard, Jakob Madsen, and Mark Wohar. 2017. Long-run commodity prices, economic growth, and interest rates:
17th century to the present day. World Development 89: 57–70. [CrossRef]
Henderson, Brian, Neil Pearson, and Li Wang. 2015. New evidence on the financialization of commodity markets. Review of Financial
Studies 28: 1285–311. [CrossRef]
Hu, Conghui, Zhibing Li, and Xiaoyu Liu. 2020. Liquidity shocks, commodity financialization, and market comovements. Journal of
Futures Markets 40: 1315–36. [CrossRef]
Huang, Jianbai, Yingli Li, Hongwei Zhang, and Jinyu Chen. 2021. The effects of uncertainty measures on commodity prices from a
time-varying perspective. International Review of Economics & Finance 71: 100–14.
Ioannidis, Christos, and Kook Ka. 2018. The impact of oil price shocks on the term structure of interest rates. Energy Economics 72:
601–20. [CrossRef]
Jahan, Sayeeda, and Apostolos Serletis. 2019. Business cycles and hydrocarbon gas liquids prices. The Journal of Economic Asymmetries
19: e00115. [CrossRef]
Kagraoka, Yusho. 2016. Common dynamic factors in driving commodity prices: Implications of a generalized dynamic factor model.
Economic Modelling 52: 609–17. [CrossRef]
Kucher, Oleg, and Suzanne McCoskey. 2017. The long-run relationship between precious metal prices and the business cycle. Quarterly
Review of Economics and Finance 65: 263–75. [CrossRef]
Kuosmanen, Petri, Jaana Rahko, and Juuso Vataja. 2019. Predictive ability of financial variables in changing economic circumstances.
North American Journal of Economics and Finance 47: 37–47. [CrossRef]
Labys, Walter, Aymad Achouch, and Michael Terraza. 1999. Metal prices and the business cycle. Resources Policy 25: 229–38. [CrossRef]
Levine, Ari, Yao Ooi, Matthew Richardson, and Caroline Sasseville. 2018. Commodities for the long run. Financial Analysts Journal 74:
55–68. [CrossRef]
McMillan, David. 2021a. Predicting GDP growth with stock and bond markets: Do they contain different information? International
Journal of Finance and Economics 26: 3651–75. [CrossRef]
McMillan, David. 2021b. When and Why Do Stock and Bond Markets Predict Economic Growth? Quarterly Review of Economics and
Finance 80: 331–43. [CrossRef]
J. Risk Financial Manag. 2021, 14, 585 39 of 39

Newey, Whitney, and Kenneth West. 1987. Hypothesis testing with efficient method of moments estimation. International Economic
Review 28: 777–87. [CrossRef]
Orlowski, Lucjan. 2017. Volatility of commodity futures prices and market-implied inflation expectations. Journal of International
Financial Markets, Institutions and Money 51: 133–41. [CrossRef]
Plosser, Charles, and Geert Rouwenhorst. 1994. International term structures and real economic growth. Journal of Monetary Economics
33: 133–55. [CrossRef]
Qadan, M., and H. Nama. 2018. Investor sentiment and the price of oil. Energy Economics 69: 42–58. [CrossRef]
Qadan, Mahmoud, David Aharon, and Ron Eichel. 2019. Seasonal patterns and calendar anomalies in the commodity market for
natural resources. Resources Policy 63: 101435. [CrossRef]
Saar, Dan, and Yossi Yagil. 2015. Forecasting growth and stock performance using government and corporate yield curves: Evidence
from the European and Asian markets. Journal of International Financial Markets, Institutions and Money 37: 27–41. [CrossRef]
Stock, James, and Mark Watson. 1989. New indexes of coincident and leading economic indicators. NBER Macroeconomics Annual 4:
351–94. [CrossRef]
Stock, James, and Mark Watson. 2003. Forecasting output and inflation: The role of asset prices. Journal of Economic Literature 41:
788–829. [CrossRef]
Tang, Ke, and Wei Xiong. 2012. Index investment and the financialization of commodities. Financial Analysts Journal 68: 54–74.
[CrossRef]
Umar, Zaghum, Syed Shahzad, and Dimitris Kenourgios. 2019. Hedging US metals & mining Industry’s credit risk with industrial and
precious metals. Resources Policy 63: 101472.
Wheelock, David, and Mark Wohar. 2009. Can the term spread predict output growth and recessions? A survey of the literature.
Federal Reserve Bank of St. Louis Review 91: 419–40. [CrossRef]