725_sample exam_2_2010

# 725_sample exam_2_2010 - ECN 725 EXERCISE 2 S.C. Ahn 1. (25...

This preview shows pages 1–2. Sign up to view the full content.

1 ECN 725 EXERCISE 2 S.C. Ahn 1. (25 pts; 5 pts. for each.) Comment on each of the following statements. No point will be given if no appropriate explanation follows. (Two or three-lines long answer would be sufficient for each question.) (1) Even if some important regressors are omitted, OLS estimators are still consistent as long as sample size is large. (2) When stochastic regressors are used, OLS estimator is useless. (3) The solution to multicollinearity is always simple. Drop some regressors. (4) I regressed y on one, x 2 and x 3 . The LM test for autocorrelation suggests that the errors in the model are serially correlated. This means that OLS estimates are useless for testing any hypothesis regarding 1 , 2 and 3 . (5) R 2 from a regression of y on a set of variables is higher than R 2 from a regression of ln(y) on the same variables. This means that y is a better dependent variable than ln(y). 2. (10 pts.) Consider the model y t = + u t , where E(u t ) = 0 and var(u t ) = 24 ot x . Here, all symbols are scalar. Observe that the model includes the constant term ( ) only, and there is no other regressor. The variable x t is nonstochastic (the x t are not ones). Assume 2 1 o . A sample of 100 observations gives us Σ t x t 4 = 10000. Let ˆ be the OLS estimator of . Compute var( ˆ ). 3. (30 pts.) Consider the model y t = 1 + 2 x t + u t , where E( u t ) = 0. A sample of 100 observations yields the following regression result: (A) y t = 0.7 + 0.3 x t , R 2 = 0.7, SSE = 910

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 11/19/2010 for the course ECON 270a taught by Professor Ahn during the Spring '10 term at ASU.

### Page1 / 3

725_sample exam_2_2010 - ECN 725 EXERCISE 2 S.C. Ahn 1. (25...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online