Probability seminar: Dual regression

29 May 2015, 2.15 PM - 29 May 2015, 3.15 PM

Sami Stouli, University of Bristol

SM3, School of Mathematics

Sami Stouli, University of Bristol

We propose an alternative (`dual regression') to the quantile regression process for the global estimation of conditional distribution functions under minimal assumptions. Dual regression provides all the interpretational power of the quantile regression process while largely avoiding the need for `rearrangement' to repair the intersecting conditional quantile surfaces that quantile regression often produces in practice. Our approach relies on a mathematical programming characterization of conditional distribution functions which, in its simplest form, provides a simultaneous estimator of location and scale parameters in a linear heteroscedastic model. Sufficient conditions for existence and uniqueness of this estimator are given, and its statistical properties are derived.

Contact information

Organisers: Marton BalazsHaeran Cho

Edit this page