Economics Seminars: Daniel Gutknecht (Frankfurt)
Daniel Gutknecht (Frankfurt)
1B6, Priory Road Complex
Title: Testing for Sample Selection
Abstract: This paper provides a unified approach for detecting sample selection in nonparametric conditional quantile and mean functions. Our testing strategy consists of a two-step procedure: the first test is an omitted predictor test with the propensity score as omitted variable. This test has power against square root-alternatives. Since, as with any omnibus test, we cannot distinguish between rejection due to genuine selection or to misspecification, but the differentiation of the two causes has implications for nonparametric (point) identification and estimation of the conditional quantile function, we run a second test to detect misspecification in case of rejection at the first stage. Using only individuals with propensity score close to one, this second test relies on an `identification at infinity' argument, but accommodates cases of irregular identification. Finally, our testing procedure does not require any parametric assumptions on the selection equation, and all our results in the quantile case hold uniformly across quantile ranks in a compact set. We apply our procedure to test for selection in log hourly wages using UK Family Expenditure Survey data.