# StreamGate - THE UNIVERSITY OF NEW S OUTH WALES I ~ <>,0 <>...

This preview shows pages 1–4. Sign up to view the full content.

, , THE UNIVERSITY OF NEW SOUTH WALES I <>, <> .,,0 <> ~ SCHOOL OF ECONOt. . IICS ECON 2206 / ECON 3290 (ARTS) INTRODUCTORY ECONO~"ETRI CS FINAL EXAM INATION SESSION 1, 2009 l. T Il\lE ALLOWED - 2 Hours. 2. REA DING TI1-.IE = 10 t. linutcs 3. T HIS EXAl\HNATlON PAPER HAS 9 PAGES 4. TOTAL NUl\IBER OF QUESTI ONS - 6. 5. ANSWER ALL QUESTIONS. 6. ALL QUESTIONS ARE OF EQUAL VALUE 7. TOTAL MAJ.lKS AVA ILABLE FOR THIS EXM,·II NATION - 60. 8. T HE l\IARKS AWARDED TO EACH PART OF A QUESTION ARE INDICATED. 9. CANDIDATES MAY BRING THEIR OWN CALCULATORS TO T HE EXAl\1 10. STATISTICA L TABLES ARE PROVIDED AT T HE END OF THE EXAt. .1 PAPER 11. ALL ANSWERS I\IUST BE WRITTEN I N PEN. PENCILS I\I AY BE USED ONLY FOR DRAWING, SKETCHI NG OR GRAPHICAL WORK. 12. THIS PAPER I\I AY BE RETAINED BY T HE CANDTDATE Please see over

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

View Full Document
ANSWER ALL SIX QUESTIONS REMINDER: When performing statistical tests, always state the null and alternative hypothe- ses, the test statistic and it's distribution under the null hypothesis, the level of significance and the conclusion of the test. Question 1. (lO Marks). (i) Consider the multiple regression model: y =fjo + PI Xl +fj'l:t2 +u Briefly explain the consequences of measurement error in the dependent variable, y, for the OLS estimators of fjl and fj2 when: (a) the measurement error is uncorrelated with the independent variables XI and X'l (2 marks) (b) the measurement error is correlated with the independent variable (ii) Consider the following model used to the salary of chief executive officers: salary = Po + 13) sales + fj2 mktva/ + fj3 sales x m1·},-val + u where sales is the firm's total sales revenue for the previous year (measured in Smillions) and 1HT},;val is the firm's market value (also measured in \$millions). III terms of the parameters of the population model, what is the effect on salary of all extra \$lmillion of sales in the previous year? (2 (iii) What is the meaning of the term "contemporaneous exogeneity" as used in the context of time series data? \Vhat is the difference between contemporancous exogeneity and the zero conditional mean (ZCM) as~IIJJlPtion ill the multiple regres~ion model for time series data ? {2 marks} (iv) What does it mean for an estimator to be "consistent" and does this differ from "ullbiasness" ? Explain. (2 marks). Please see over 2 I
Question 2. (lO Marks in total) The variable heavyd is a binary variable equal to one if a person is a heavy drinker (of alcoho\), and zero otherwise. T he estimates for the linear probability model explaining was: = 0.467 - 0.007 log(price) + 0.056 log(i71come) + 0.028educ - 0.000geduc 2 - 0.014 age (0.210) (0.034) (0.029) (0.006) (0.0003) (0.06) [0.131[ [0.031[ [O.OI B[ [0.005] [0.0002] [0.05] 71 = 1403, n2 = 0.233 where price is the price of beer in the individual's local area, income is monthly income, educ is years of education and age is measured ill years. The usual OLS standard errors are reported in parentheses (.), and the heteroskedasticity-robust standard errors are reported in brackets [.]. (i) At what point (i.e. specific number of years) does anot her year of education reduce the probability of being a heavy drinker ? (l mark) (ii) Are there any important differences between the two sets of standard errors reported above? Briefly explain. (2

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.

{[ snackBarMessage ]}

### Page1 / 9

StreamGate - THE UNIVERSITY OF NEW S OUTH WALES I ~ <>,0 <>...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online