Stepwise Choice of Covariates in High Dimensional Regression
Methodology
2017-10-06 v4
Abstract
Given data y(n) and p(n)covariates x(n) one problem in linear regression is to decide which if any of the covariates to include. There are many articles on this problem but all are based on a stochastic model for the data. This paper gives what seems to be a new approach which does not require any form of model. It is conceptually and algorithmically simple and consistency results can be proved under appropriate assumptions.
Keywords
Cite
@article{arxiv.1610.05131,
title = {Stepwise Choice of Covariates in High Dimensional Regression},
author = {Laurie Davies},
journal= {arXiv preprint arXiv:1610.05131},
year = {2017}
}
Comments
This is a revised version of 1610.05131. It contains some results on false postives, an analysis of the birthday data also analysed in "Bayesian Data Analysis" (Chapman & Hall/CRC Texts in Statistical Science) and an application to the construction of dependency graphs. 38 pages and one figure