Inference for Extremal Conditional Quantile Models, with an Application to Market and Birthweight Risks
为极值条件分位数模型提供了可行的推断工具，利用极值近似自归一化分位数回归统计量的分布，方法简单且适用于非回归情形，并通过股票收益极端波动和婴儿出生体重极低百分位两个实例展示应用。
Quantile regression is an increasingly important empirical tool in economics\nand other sciences for analyzing the impact of a set of regressors on the\nconditional distribution of an outcome. Extremal quantile regression, or\nquantile regression applied to the tails, is of interest in many economic and\nfinancial applications, such as conditional value-at-risk, production\nefficiency, and adjustment bands in (S,s) models. In this paper we provide\nfeasible inference tools for extremal conditional quantile models that rely\nupon extreme value approximations to the distribution of self-normalized\nquantile regression statistics. The methods are simple to implement and can be\nof independent interest even in the non-regression case. We illustrate the\nresults with two empirical examples analyzing extreme fluctuations of a stock\nreturn and extremely low percentiles of live infants' birthweights in the range\nbetween 250 and 1500 grams.\n