Unbiasedness in Least Quantile Regression

DefaultRisk.com the web's biggest credit risk modeling resource.

Submit Your Paper

Post Your Résumé

For Recruiters

Today's Featured Book

The Handbook of Portfolio Mathematics: Formulas for Optimal Allocation & Leverage
by Ralph Vince, Wiley, (May 25, 2007), Hardcover, 448 pages

Training Discounted for DefaultRisk.com visitors only:

The Mathematics of Credit Derivatives: The Essential Credit Modelling and Pricing Companion
by Philipp J. Schönbucher,
WBS Training, August 2003, DVD / Interactive CD-ROM

Sponsor:
Shop at Amazon.com and support DefaultRisk.com

Unbiasedness in Least Quantile Regression

by Dirk Tasche of the Technische Universität München

September 6, 2001

Abstract: We develop an abstract notion of regression which allows for a non-parametric formulation of unbiasedness. We prove then that least quantile regression is unbiased in this sense even in the heteroscedastic case if the error distribution has a continuous, symmetric, and uni-modal density. An example shows that unbiasedness may break down even for smooth and symmetric but not uni-modal error distributions. We compare these results to those for least MAD and least squares regression.

Keywords: Least Quantile, Regression, Unbiasedness, Fisher consistency, Quantile Derivative, Lord's paradox.

Books Referenced in this Paper: (what is this?)