A1262
Title: Lambda quantile regression and its application to finance
Authors: Ilaria Peri - Birkbeck-University of London (United Kingdom) [presenting]
Abstract: A rigorous foundation of Lambda-quantile regression is provided. Lambda-quantiles have been introduced as a generalization of the notion of quantiles which maintain the basic structure of the usual definition but add more flexibility in the choice of the threshold lambda. We generalize the classical quantile regression in an extended framework that encompasses both conditional quantiles and Lambda quantiles. In particular, we focus on the linear case in analogy with the usual quantile regression. We conduct an empirical exercise where we compare two different strategies for forecasting Lambda quantiles of the Standard \& Poor's 500 index. Finally, we present the results of the backtesting experiment.
