A group of researchers wishes to test whether the volume of industrial production in

logarithm function over time (lip) affect return on the All-Ordinaries index (a trading
index in Australia, illustrating the power of heavy industry in Austrialia from top 50
companies) in logarithm function (lallords). Use Appendix A to answer the following
questions.

(see APPENDIX A output for the questions in this SECTION)

a) Discuss the characteristics of the data by making reference to the lower subscript.
Write the regression equation

b) What signs would you expect on Beta 1? Why?

c) Report the least squares estimates of the coefficient and corresponding p-value.

d) Are the sign of the estimated coefficient Beta1 in accord with economic theory? Give
the interpretation to the estimates for Beta1.

e) Determine whether the estimated coefficients are statistically significant at the 5%

level of significance.

f) Test for the overall significance of the model at 5% level of significance. Interpret the
meaning of the coefficient of determination.

g) Conduct test on the model to check whether it is omitted or not and discuss your
results.

h) Check whether the error term in the model is normally distributed or not and explain
your answer.

i) Conduct diagnostic tests to check whether the model satisfies the assumptions of
OLS and discuss your results.

j) Conduct Stationary test on the variable lallords, lip and discuss your answer

k) Are the two variables lallords and lip have the same unit root with four lags ? Are they
cointergrated ?
l) The group of researchers now are interested to know whether the volume of industrial
production in logarithm function over time (lip) affect return on the All-Ordinaries index
(a trading index in Australia, illustrating the power of heavy industry in Austrialia from top
50 companies) in logarithm function different to what extent in the short-term and long-
term ? Is this regression a spurious regression ? Write the two models of short and long-
term ? How the model in short-term could be used to explain the long-term?
 Tuesday December 5 17:43:01 2023 Page 1

___ ____ ____ ____ ____(R)

/__ / ____/ / ____/
___/ / /___/ / /___/
Statistics/Data analysis

User: Intergrated Problem

___ ____ ____ ____ ____ ©

/__ / ____/ / ____/ 17.0
___/ / /___/ / /___/ MP—Parallel Edition

Statistics and Data Science Copyright 1985-2021 StataCorp LLC

StataCorp
4905 Lakeway Drive
College Station, Texas 77845 USA
800-STATA-PC https://www.stata.com
979-696-4600 [email protected]

Stata license: Unlimited-user 64-core network perpetual

Serial number: 18461036
Licensed to: TEAM BTCR
TEAM BTCR

Notes:
1. Unicode is supported; see help unicode_advice.
2. More than 2 billion observations are allowed; see help obs_advice.
3. Maximum number of variables is set to 5,000; see help set_maxvar.
4. New update available; type -update all-

1 . use "C:\Users\VINH LUAN\OneDrive\Máy tính\[Aim 90] EFM\EFM Notes\Module 9_Modelling Highly Persistent Time Seri
> gration\ausvolat.dta"

2 . format date %td

3 .
4 . generate qtr = quarter(date)

5 .
6 . generate yr = year(date)

7 .
8 . generate datestring = string(yr) + "q" + string(qtr)

9 .
10 . generate qtryr = quarterly(datestring, "YQ")

11 .
12 . format qtryr %tq

13 .
14 . tsset qtryr, quarterly

Time variable: qtryr, 1996q1 to 2015q4

Delta: 1 quarter

15 .
16 . sort qtryr
 Tuesday December 5 17:43:01 2023 Page 2

17 .
18 . generate t = _n

19 .
20 . drop qtr datestring yr

21 . sum lallords lip t

Variable Obs Mean Std. dev. Min Max

lallords 77 8.255316 .2878485 7.717114 8.795507

lip 77 4.452492 .1154396 4.25703 4.67935
t 80 40.5 23.2379 1 80

22 . reg lallords lip t

Source SS df MS Number of obs = 77

F(2, 74) = 102.53
Model 4.62731295 2 2.31365647 Prob > F = 0.0000
Residual 1.66980068 74 .022564874 R-squared = 0.7348
Adj R-squared = 0.7277
Total 6.29711363 76 .082856758 Root MSE = .15022

lallords Coefficient Std. err. t P>|t| [95% conf. interval]

lip -.1566272 1.034749 -0.15 0.880 -2.21841 1.905155

t .0118286 .0053393 2.22 0.030 .0011898 .0224675
_cons 8.491381 4.401291 1.93 0.058 -.2783829 17.26115

23 . estat ovtest

Ramsey RESET test for omitted variables

Omitted: Powers of fitted values of lallords

H0: Model has no omitted variables

F(3, 71) = 9.67

Prob > F = 0.0000

24 . predict u, resid
(3 missing values generated)

25 . sktest u

Skewness and kurtosis tests for normality

Joint test
Variable Obs Pr(skewness) Pr(kurtosis) Adj chi2(2) Prob>chi2

u 77 0.0001 0.0409 14.88 0.0006

 Tuesday December 5 17:43:01 2023 Page 3

26 . estat vif

Variable VIF 1/VIF

lip 48.06 0.020808

t 48.06 0.020808

Mean VIF 48.06

27 . hettest u

Breusch–Pagan/Cook–Weisberg test for heteroskedasticity

Assumption: Normal error terms
Variable: u

H0: Constant variance

chi2(1) = 53.35
Prob > chi2 = 0.0000

28 . reg lallords lip t

Source SS df MS Number of obs = 77

F(2, 74) = 102.53
Model 4.62731295 2 2.31365647 Prob > F = 0.0000
Residual 1.66980068 74 .022564874 R-squared = 0.7348
Adj R-squared = 0.7277
Total 6.29711363 76 .082856758 Root MSE = .15022

lallords Coefficient Std. err. t P>|t| [95% conf. interval]

lip -.1566272 1.034749 -0.15 0.880 -2.21841 1.905155

t .0118286 .0053393 2.22 0.030 .0011898 .0224675
_cons 8.491381 4.401291 1.93 0.058 -.2783829 17.26115

29 . estat imtest

Cameron & Trivedi's decomposition of IM-test

Source chi2 df p

Heteroskedasticity 13.42 5 0.0197

Skewness 10.66 2 0.0048
Kurtosis 4.52 1 0.0335

Total 28.60 8 0.0004

 Tuesday December 5 17:43:01 2023 Page 4

30 . dfuller lallords, lags(4) reg

Augmented Dickey–Fuller test for unit root

Variable: lallords Number of obs = 72

Number of lags = 4

H0: Random walk without drift, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) -1.418 -3.549 -2.912 -2.591

MacKinnon approximate p-value for Z(t) = 0.5734.

Regression table

D.lallords Coefficient Std. err. t P>|t| [95% conf. interval]

lallords
L1. -.0397711 .0280395 -1.42 0.161 -.0957537 .0162115
LD. .3638756 .1215327 2.99 0.004 .1212278 .6065235
L2D. -.1327762 .1291044 -1.03 0.307 -.3905415 .124989
L3D. .0788414 .1285248 0.61 0.542 -.1777667 .3354494
L4D. -.0966476 .1215317 -0.80 0.429 -.3392936 .1459983

_cons .3389742 .2318126 1.46 0.148 -.1238546 .8018031

31 . dfuller lip, lags(4) reg

Augmented Dickey–Fuller test for unit root

Variable: lip Number of obs = 72

Number of lags = 4

H0: Random walk without drift, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) 0.367 -3.549 -2.912 -2.591

MacKinnon approximate p-value for Z(t) = 0.9802.

Regression table

D.lip Coefficient Std. err. t P>|t| [95% conf. interval]

lip
L1. .0045166 .012299 0.37 0.715 -.0200392 .0290725
LD. -.0200633 .1182433 -0.17 0.866 -.2561437 .2160171
L2D. -.0530287 .1108791 -0.48 0.634 -.274406 .1683487
L3D. -.3841043 .1109286 -3.46 0.001 -.6055804 -.1626281
L4D. -.1541981 .1168453 -1.32 0.192 -.3874872 .0790911

_cons -.0109658 .0545391 -0.20 0.841 -.1198567 .0979251

 Tuesday December 5 17:43:01 2023 Page 5

32 . dfuller d1.lallords, lags(4) reg

Augmented Dickey–Fuller test for unit root

Variable: D.lallords Number of obs = 71

Number of lags = 4

H0: Random walk without drift, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) -3.780 -3.551 -2.913 -2.592

MacKinnon approximate p-value for Z(t) = 0.0031.

Regression table

D2.lallords Coefficient Std. err. t P>|t| [95% conf. interval]

lallords
LD. -.8314755 .2199723 -3.78 0.000 -1.270791 -.3921604
LD2. .1943364 .1955625 0.99 0.324 -.1962288 .5849016
L2D2. .0296667 .1755213 0.17 0.866 -.3208736 .3802069
L3D2. .1095696 .145894 0.75 0.455 -.1818009 .40094
L4D2. -.0108275 .1234492 -0.09 0.930 -.2573726 .2357177

_cons .0093729 .0078682 1.19 0.238 -.006341 .0250869

33 . dfuller d1.lip, lags(4) reg

Augmented Dickey–Fuller test for unit root

Variable: D.lip Number of obs = 71

Number of lags = 4

H0: Random walk without drift, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) -5.294 -3.551 -2.913 -2.592

MacKinnon approximate p-value for Z(t) = 0.0000.

Regression table

D2.lip Coefficient Std. err. t P>|t| [95% conf. interval]

lip
LD. -1.66263 .3140399 -5.29 0.000 -2.289811 -1.035449
LD2. .665438 .2546017 2.61 0.011 .1569632 1.173913
L2D2. .5893883 .2043491 2.88 0.005 .1812749 .9975017
L3D2. .2049245 .1698184 1.21 0.232 -.1342263 .5440753
L4D2. .0850217 .1186085 0.72 0.476 -.1518558 .3218992

_cons .0092868 .0021399 4.34 0.000 .0050131 .0135605

 Tuesday December 5 17:43:01 2023 Page 6

34 . predict ut, resid

(3 missing values generated)

35 . dfuller ut, noconstant lags(4) reg

Augmented Dickey–Fuller test for unit root

Variable: ut Number of obs = 72

Number of lags = 4

H0: Random walk without drift, a = 0, d = 0

Dickey–Fuller
Test critical value
statistic 1% 5% 10%

Z(t) -2.438 -2.611 -1.950 -1.610

Regression table

D.ut Coefficient Std. err. t P>|t| [95% conf. interval]

ut
L1. -.1330067 .054555 -2.44 0.017 -.241899 -.0241144
LD. .4055183 .1186093 3.42 0.001 .1687732 .6422635
L2D. -.0745267 .12802 -0.58 0.562 -.3300557 .1810023
L3D. .1253127 .1258995 1.00 0.323 -.1259838 .3766093
L4D. -.0330276 .122035 -0.27 0.787 -.2766104 .2105552

36 . reg d1.lallords d1.lip t l1.u

Source SS df MS Number of obs = 76

F(3, 72) = 1.89
Model .021684469 3 .007228156 Prob > F = 0.1394
Residual .275740919 72 .003829735 R-squared = 0.0729
Adj R-squared = 0.0343
Total .297425389 75 .003965672 Root MSE = .06188

D.lallords Coefficient Std. err. t P>|t| [95% conf. interval]

lip
D1. .7887486 .621158 1.27 0.208 -.4495073 2.027005

t -.0001691 .0003251 -0.52 0.605 -.0008172 .0004791

u
L1. -.1008611 .0481582 -2.09 0.040 -.1968626 -.0048595

_cons .0147689 .0147663 1.00 0.321 -.0146672 .0442051

37 .