AN ALTERNATIVE ESTIMATOR FOR THE CENSORED QUANTILE REGRESSION-MODEL

The paper introduces an alternative estimator for the linear censored quantile regression model. The objective function is globally convex a nd the estimator is a solution to a linear programming problem. Hence, a global minimizer is obtained in a finite number of simplex iteratio ns. The suggested estimator also applies to the case where the censori ng point is an unknown function of a set of regressors. It is shown th at, under fairly weak conditions, the estimator has a root n-convergen ce rate and is asymptotically normal. In the case of a fixed censoring point, its asymptotic property is nearly equivalent to that of the es timator suggested by Powell (1984, 1986a). A Monte Carlo study perform ed shows that the suggested estimator has very desirable small sample properties. It precisely corrects for the bias induced by censoring, e ven when there is a large amount of censoring, and for relatively smal l sample sizes.