Weighted Linear Quantile Regression In LTRC Data Model

In many prospective and retrospective studies,survive data are subject to left truncation in addition to the usual right censoring. Quantile regression offers graet flexibility in assessing covariate efforts on the response.In this article, based on the Weihua Zhou(2010),we develop a quantile regression for left-truncated and right-censored data.Firstly, we introduce the regression model,the mean-ing of the parameters and the detailed definition of the LTRC data. Secondly in LTRC model we discuss the identifiable condition, under which we derive the estimators of all distribution functions with product-limit estimator. A weighted quantile regression for LTRC data is obtained with the method of local linear quantile regression, so is the the specific form of the objective function. Finally, we evaluate the finite sample performance of the proposed estimators under the homoscedastic and the heteroscedastic model.And we get the estimators of the regression coefficient β under different quantiles.