DEVELOPMENT OF A CROSS-SECTIONALLY CONSTRAINED RANDOM COEFFICIENT MODEL, WITH AN APPLICATION TO THE ESTIMATION OF THE VINTAGE EFFECT ON EARNINGS PROFILES

LINDA M. PIUNNO WOODLAND, Purdue University

Abstract

This research extends the random coefficient model in a seemingly unrelated regressions framework, developed originally by Singh and Ullah. The model imposes constraints on the mean response coefficients for various homogeneous groups of individuals of different cross sections and develops estimation procedures under various error structures. Statistical tests for relationships of coefficients between groups are also derived. The model is applied to the estimation of the effect on an individual's vintage in his earnings over his lifetime, using panel data for Ph.D. scientists. A hypothesis test is performed which verifies that the vintage effect exists.

Degree

Ph.D.

Subject Area

Economic theory

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS