Applied Quantile Regression (303094)

Keywords: quantile regression, conditional distribution estimation, conditional ranking

Quantile regression is a modern statistical methodology for modeling quantiles of a response variable conditional on explanatory covariates. Whereas linear regression models the conditional mean, quantile regression enables you to more fully explore your data by modeling the conditional quantiles, such as the median and the 5th and 95th percentiles. Quantile regression is particularly useful when your data are heterogeneous, or when you cannot assume a parametric distribution for the response. Quantile process regression fits a model for the entire quantile level range in [0,1], enabling you to estimate the entire conditional distribution of a response variable. This presentation provides an overview and several intuitive examples of the quantile regression methodology. Common application areas include market analysis, economics, environmental studies, and health science.