Abstract
The justification of applying a random coefficient regression model in econometric work has been discussed by numerous econometricians. Hildreth and Houck have derived a set of consistent estimators for such a model. Alternatives to these are developed but analytic attempts to ascertain the small sampling properties of these alternative estimators have not been very successful so far. A Monte Carlo experiment is made and the relative performance of these estimators is described.

This publication has 6 references indexed in Scilit: