Online Program Home
  My Program

Abstract Details

Activity Number: 569 - Recent Advances in Predictive Inference
Type: Topic Contributed
Date/Time: Wednesday, August 2, 2017 : 2:00 PM to 3:50 PM
Sponsor: IMS
Abstract #324153
Title: Predictive Inference for Regression
Author(s): Maxwell Grazier G'Sell*
Companies: Carnegie Mellon University
Keywords: conformal ; predictive inference
Abstract:

We develop a general framework based on conformal inference for conducting distribution-free predictive inference in regression. The proposed methodology allows construction of prediction bands for the response variable using any estimator of the regression function. The resulting prediction band preserves the consistency properties of the original estimator under standard assumptions, while guaranteeing finite sample marginal coverage even when the assumptions do not hold. We analyze and compare both full sample and split sample versions of the methods, as well as extensions to in-sample prediction and variable importance.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2017 program

 
 
Copyright © American Statistical Association