th2-642-08 - Second take home exercise for Econ 642 April...

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

View Full Document Right Arrow Icon
Second take home exercise for Econ 642 April 2008 R. de Jong 1. Consider the following dataset: we have panel data on 500 children whose parents divorced at age 8. For every year between ages 9 and 18, we have a measure of their psychologi- cal health y it (as measured by a standardized test) and a dummy variable x it that indicates whether or not child i received counseling in year t - 1. We are interested in the causal effect of counseling on psychological health. Someone runs a random effects regression of y it on x it . Why might a fixed effects regression be more appropriate in this setting? (Note: I am interested in the fundamental omitted vari- able problem with this random effects regression; the assumption is that a baseline measure of psychological health of child i , i.e. a i , is the omitted variable) 2. In the setup of the previous question, what results would you expect to obtain from a pooled regression?
Background image of page 1

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

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 07/17/2008 for the course ECON 642 taught by Professor De jong during the Spring '08 term at Ohio State.

Page1 / 4

th2-642-08 - Second take home exercise for Econ 642 April...

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