# ECON 5420 Study Guide - Final Guide: Cointegration, Autocorrelation, Unit Root

46 views6 pages

Published on 5 Oct 2018

School

Department

Course

Professor

Exam 1 Econ 5420, Econometrics II, Fall 2017

Section: Name:

1 Multiple Choice

Choose the answer that best answers the question. Each question is worth 3 points.

Answers must be clearly written on the supplied answer sheet.

1. For the time series xt=ρxt−1+etto be covariance stationary, what must be true?

(a) ρ= 1

(b) ρ > 1

(c) ρ < 1

(d) None of the above.

2. Given the model yt=α+δ0xt+δ1xt−1+δ2xt−2+et, which describes the short

run-propensity?

(a) α

(b) α+δ0

(c) δ0

(d) δ1

3. Given ytis I(2), which statement is true?

(a) ∆ytis stationary

(b) yt−1is stationary

(c) ytis non-stationary

(d) None of the above

4. If the equation yt=α+βxt+etis estimated with heteroskedasticity, what are

the consequence(s) for ˆ

β:

(a) standard errors are valid, whereas the t statistics and F statistics are invalid.

(b) ˆ

βis biased

(c) ˆ

βis unbiased, but not consistent

(d) standard errors, t statistics, and F statistics are invalid.

https://www.coursehero.com/file/33915519/Time-Series-Exam-f17pdf/

5. Which of the following is required to ﬁnd unbiased estimates of ˆ

β1in the equation:

yt=β0+β1xt+et

(a) E[ut|xt, xt−1, xt−2, ...] = 0

(b) No serial correlation.

(c) No heteroskedasticity.

(d) All of the above

6. Which of the following will bias ˆ

β1in the equation yt=β0+β1xt+et?

(a) ythas a unit root.

(b) ytand xthave time trends

(c) ytand xthave seasonal trends

(d) All of the above

7. Which of the following describes the critical values for a t-distribution (tc), Dickey

Fuller (DFc) distribution, and Engle-Granger test for cointegration (EGc)?

(a) |tc|<|DFc|=|EGc|

(b) |tc|>|DFc|=|EGc|

(c) |tc|>|DFc|>|EGc|

(d) |tc|<|DFc|<|EGc|

8. Which of the following can correct for both serial correlation and heteroskedas-

ticity?

(a) quasi-diﬀerencing

(b) Cochrane-Orcutt estimation

(c) Newey West HAC standard errors

(d) Durbin-Watson Test

9. Which is true of a ﬁnite distributed lag model?

(a) The model is dynamic.

(b) The model is static.

(c) All variables are from the same time period.

(d) All of the above

https://www.coursehero.com/file/33915519/Time-Series-Exam-f17pdf/