This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: ECON 140, Fall 2008 Alex Rothenberg Practice Problems: Multiple Regression, Regression on Indicators, Interactions Problem 1 This problem asks you to look at the results of a famous (and controversial) study: Donohue, J. and S. Levitt (2001) “The Impact of Legalized Abortion on Crime”, Quarterly Journal of Economics 66(2) pp. 379-420. The authors estimate the parameters of the following regression equation using data from all 50 states (plus D.C.), indexed by s , and years, t = 1985- 1997. ln CRIME st = β ABORT st + K X k =1 θ k X k,st + γ s + η t + ε st Table 3 presents summary statistics for the variables used in estimating this equation, and Table 4 presents the regression results. The variable ABORT st is called “effective abortions”, essentially coded as the number of abortions in state s at year t- 18. 1 The authors do some fancy things when estimating this equation, but we’re going to assume that it was estimated using OLS. The number of observations in each equation isusing OLS....
View Full Document
This document was uploaded on 02/07/2010.
- Spring '09