MISCLASSIFICATION OF THE DEPENDENT VARIABLE IN A DISCRETE-RESPONSE SETTING

Misclassification of dependent variables in a discrete-response model causes inconsistent coefficient estimates when traditional estimation techniques (e.g., probit or legit) are used. A modified maximum likeli hood estimator that corrects for misclassification is proposed. A semi parametric approach, which combines the maximum rank correlation estim ator of Han (1987) (Journal of Econometrics 35, 303-316) with isotonic regression, allows for more general forms of misclassification than t he maximum likelihood approach. The parametric and semiparametric esti mation techniques are applied to a model of job change with two common ly used data sets, the Current Population Survey (CPS) and the Panel S tudy of Income Dynamics (PSID). (C) 1998 Elsevier Science S.A. All rig hts reserved.