Section_5_Spring2008

Section_5_Spring2008 - 02 Educ 14 145 12(0 03(0 03(0 03...

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

View Full Document Right Arrow Icon
Practice Problem: F-test and OVB Econ 142 Spring 2008 Raymundo M. Campos-Vazquez February 20, 2008 Contents 1 Problem 1: F-tests 1 1 Problem 1: F-tests You want to estimate the causal e/ect of smoking on wages. You estimate the following regression: ln w = & 0 + 1 Smoke + 2 Age + 3 Education + u Smoke is a dummy variable for smoking, Age is the age of indvidual, Education is the years more regression models. One including a dummy variable for female, and the last model includes a dummy variable for female, age squared and an interaction term between age and education. The following table includes the parameters from the three regressions (Note: Numbers Column (2) includes female and (3) include female and other terms. Standard errors are shown in parentheses: 1
Background image of page 1

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

View Full Document Right Arrow Icon
(1) (2) (3) Smoke -0.10 -0.12 -0.11 (0 : 03) (0 : 01) (0 : 01) Age 0 : 05 0 : 05 0 : 10 (0 : 01) (0 : 01) (0
Background image of page 2
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: : 02) Educ : 14 : 145 : 12 (0 : 03) (0 : 03) (0 : 03) Female & : 30 & : 32 (0 : 10) (0 : 10) Age 2 & : 05 (0 : 03) Age ± Educ : 10 (0 : 04) R 2 : 50 : 53 : 60 R 2 : 48 : 51 : 57 n 100 100 100 1. Using Column (1), what is the e/ect of smoking on wages? 2. Under which conditions would the coe¢ cient on smoking give you a causal interpreta-tion of smoking on wages? 3. Test the null hypothesis that the population parameter of smoking is equal to -0.15 agianst the alternative the paramter is di/erent from -0.15. 4. Using Column (1), test the signi&cance of the regression (i.e. test whether all parame-ters but the constant are equal to zero). 5. Comparing Column (1) and (2) you observe that the coe¢ cient on smoking is more negative in column (2) than in column (1). Why is this the case? 6. Test whether you should use Model (3) instead of Model 1 at the 5% signi&cance level. 2...
View Full Document

{[ snackBarMessage ]}

Page1 / 2

Section_5_Spring2008 - 02 Educ 14 145 12(0 03(0 03(0 03...

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

View Full Document Right Arrow Icon
Ask a homework question - tutors are online