RECENT ADVANCES IN QUANTILE REGRESSION-MODELS - A PRACTICAL GUIDELINEFOR EMPIRICAL-RESEARCH

This paper provides a guideline for the practical use of the semi-para metric technique of quantile regression, concentrating on cross-sectio n applications. It summaries the most important issues in quantile reg ression applications and fills some gaps in the literature. The paper (a) presents several alternative estimators for the covariance matrix of the quantile repression estimates: (b) reviews the results for a se quence of quantile regression estimates; and (c) discusses testing pro cedures for homoskedasticity and symmetry of the error distribution. T he various results in the literature are incorporated into the general ized method of moments framework. The paper also provides art empirica l example rising data from the Current Population Survey, raising seve ral important issues relevant to empirical applications of quantile re gression. The paper concludes with an extension to the censored quanti le regression model.