Title

Kernel Resampling To Improve The Performance Of Multiple Regression With Small Samples

Keywords

Bootstrap; Estimation bias; Kernel method; Multiple regression; Resampling; Smoothing; Standard error

Abstract

The issues of estimation accuracy and statistical power in multiple regression with small samples have long been a concern. The present study utilizes a new multivariate resampling method, the kernel resampling technique (KRT), to improve the estimation accuracy and statistical power in multiple regression with small samples. KRT is a distribution-free method that employs kernel technique to create multivariate resamples based on a given small sample. The findings from both a simulation study and an empirical example suggest that the statistical performance of multiple regression has been improved through KRT. © 2010 Nova Science Publishers, Inc.

Publication Date

12-28-2010

Publication Title

Journal of Applied Statistical Science

Volume

17

Issue

4

Number of Pages

573-582

Document Type

Article

Personal Identifier

scopus

Socpus ID

78650430635 (Scopus)

Source API URL

https://api.elsevier.com/content/abstract/scopus_id/78650430635

This document is currently not available here.

Share

COinS