Quantile regression (QR) is a statistical technique that allows to estimate conditional quantile functions (e.g., the conditional median function) and obtain statistical inference about them in much the same way as classical regression methods based on minimizing sums of residuals facilitate estimation of conditional mean functions.
This chapter helps you to understand the main principles of quantile regression and demonstrates how to use XploRe for quantile regression analysis. As it is impossible to provide a profound introduction into this area here, we refer readers for further information to bibliography.
Before proceeding to the next section, please type at the XploRe command line
library("metrics")to load the necessary quantlibs (libraries). Quantlib metrics automatically loads xplore , kernel , glm , and multi quantlibs.