[Ml-stat-talks] Wilks Seminar: Lauren Hannah, March 8 at 12:30 PM, Sherrerd Hall 101
Ahmet Emre Barut
abarut at Princeton.EDU
Thu Mar 7 13:57:19 EST 2013
=== ORFE - Wilks Statistics Seminar Announcement ===
DATE: Friday, March 8, 2013
LOCATION: Sherrerd Hall 101
SPEAKER: Lauren Hannah, Department of Statistics, Columbia University
TITLE: Bayesian Nonparametric Multivariate Convex Regression
ABSTRACT: In many applications, such as economics, operations research and reinforcement learning, one often needs to estimate a multivariate regression function subject to a convexity constraint. For example, in economics a production function may be known to be convex or concave. We propose a new Bayesian nonparametric multivariate approach based on characterizing the unknown regression function as the max of a random collection of unknown hyperplanes. We show asymptotic adaptive convergence rates for this method and design an efficient reversible jump Markov chain Monte Carlo algorithm for posterior computation. We apply our method to predicting unemployment rates based on age and years of experience using the 2012 Current Population Survey data set.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Ml-stat-talks