Scribd
Upload
30 views4 pages

Qaqb4 Solutions

This document contains 12 multiple choice questions related to time series modeling and forecasting techniques. Specifically, it covers topics like model selection criteria such as AIC, SIC, and MSE (questions 1-3), characteristics of stationary and white noise processes (questions 4-6), lag operators and autocorrelation functions (questions 7-9), identifying AR, MA, and ARMA processes based on autocorrelation patterns (questions 10-12). The questions test understanding of key concepts and properties of different time series models.

Uploaded by

askshreyasbdavey
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
30 views4 pages

Qaqb4 Solutions

This document contains 12 multiple choice questions related to time series modeling and forecasting techniques. Specifically, it covers topics like model selection criteria such as AIC, SIC, and MSE (questions 1-3), characteristics of stationary and white noise processes (questions 4-6), lag operators and autocorrelation functions (questions 7-9), identifying AR, MA, and ARMA processes based on autocorrelation patterns (questions 10-12). The questions test understanding of key concepts and properties of different time series models.

Uploaded by

askshreyasbdavey
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Modeling and Forecasting Trend

Q1. If the lines below represent s2, AIC and SIC for the same sample size, the AIC and SIC respectively are
represented by:

A. Blue and Red

B. Green and Red

C. Blue and Green

D. Red and Green

Ans: B, SIC has the largest penalty therefore steepest, followed by AIC which has a bigger penalty than s2

Q2. Which of the following is the most consistent criteria for selecting a model:

A. s2

B. SIC

C. AIC

D. MSE

Ans: B, Of the four options SIC is most consistent as it has the largest penalty for degrees of freedom.

Q3. Which of the following represents the penalty factor according to SIC:

A.

B.

C.
D. 1

Ans: C

Characterizing Cycles
Q4. Which of the following is not a necessary condition for a time series to be covariance stationary?
A. Constant and finite expected value

B. Constant and finite variance

C. Zero mean

D. Constant and finite covariance between values at a given time lag

Ans: C, The mean for a covariance stationary process needs to be constant and not necessarily zero

Q5. Which of the following is not consistent with white noise?

A. Constant mean

B. Constant variance

C. Finite variance

D. No serial correlation

Ans: A, the mean for a white noise process needs to be zero

Q6. Which of the following Q-statistics work better with limited data?

A. Box-Pierce

B. Ljung-Box

C. Both Box-Pierce and Ljung-Box are equally good

D. Neither is useful

Ans: B, by incorporating weights Liung-Box has a better chance of picking the right model compared to
the Box-Pierce method

Q7. A lag operator doesn’t:

A. Help evolve the relationship between lagged data and future data

B. Use only in-sample data to estimate

C. Consider only finite order polynomials

D. Use past, current and future values

Ans: D, lag operators use past and present data that is insample to estimate out of sample future data

Modeling Cycles

Q8. The order of the MA process with the following data is:
Lag Autocorrelation t-Statistic

1 1.4609 –6.8912

2 1.4384 5.4589

3 1.4589 6.1204

4 0.9875 –6.2345

5 0.0356 0.0132

A. 2

B. 3

C. 4

D. 5

Ans: C, Cutoff happens at the degree at which the mean autocorrelation becomes statistically
insignificant

Q9. Which of the following processes don’t show autocorrelation cutoff?

A. MA process only

B. AR and ARMA processes

C. MA and ARMA processes

D. AR, MA and ARMA processes

Ans: B, MA processes always exhibit a autocorrelation cutoff corresponding to the degree of the model

Q10. Which of the following represents an AR process?

A.
B.
C.
D.
Ans: C

Q11. What is the order of the following process?


A. 1

B. 2

C. 4

D. 5

Ans: B, order is defined by number of lagged values considered and not the degree of lag

Q12. Which of the following statements regarding ARMA process are not true?

A. The autocorrelations for the data cutoff after a point

B. The autocorrelations can exhibit periodic spikes

C. The autocorrelations show a consistent gradual decay and don’t cutoff

D. The ARMA process doesn’t consider the lagged values of the independent variable

Ans: B, There could be periodic spikes in autocorrelations because of the MA component of the process.
ARMA considers lagged values of the dependent variable

You might also like