Misspecification Testing in a Class of Conditional Distributional Models

38 Pages Posted: 25 Feb 2012

See all articles by Christoph Rothe

Christoph Rothe

Columbia University

Dominik Wied

University of Cologne

Abstract

We propose a specification test for a wide range of parametric models for the conditional distribution function of an outcome variable given a vector of covariates. The test is based on the Cramer-von Mises distance between an unrestricted estimate of the joint distribution function of the data, and a restricted estimate that imposes the structure implied by the model. The procedure is straightforward to implement, is consistent against fixed alternatives, has non-trivial power against local deviations of order n^-1/2 from the null hypothesis, and does not require the choice of smoothing parameters. In an empirical application, we use our test to study the validity of various models for the conditional distribution of wages in the US.

Keywords: Cramer-von Mises distance, quantile regression, distributional regression, location-scale model, bootstrap, wage distribution

JEL Classification: C12, C14, C31, C52, J31

Suggested Citation

Rothe, Christoph and Wied, Dominik, Misspecification Testing in a Class of Conditional Distributional Models. IZA Discussion Paper No. 6364, Available at SSRN: https://ssrn.com/abstract=2010979 or http://dx.doi.org/10.2139/ssrn.2010979

Christoph Rothe (Contact Author)

Columbia University ( email )

Dominik Wied

University of Cologne ( email )

Albertus-Magnus-Platz
Cologne, 50923
Germany

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
79
Abstract Views
986
Rank
555,299
PlumX Metrics