Robust statistical modeling using the t distribution.

Journal of the American Statistical Association, December, 1989 by Lange, Kenneth L.; Little, Roderick J.A.; Taylor, Jeremy, M.G.

Robust Statistical Modeling Using the t Distribution

1. INTRODUCTION

Statistical inference based on the normal distribution (univariate or multivariate) is known to be vulnerable to outliers. Despite this fact and the considerable interest in robust procedures in the mathematical statistical literature, most applied statistical analysis continues to be based on the normal model. Even in linear regression, where robustness concerns have penetrated statistical software widely available to practitioners, procedures are mainly directed at detecting outliers. For example, see the regression diagnostic procedures in BMDP (Dixon 1983), SAS (1982), or SPSS (1983). After editing outliers, subsequent analysis is often still restricted to least squares based on...

Read the rest of this article with a Free Trial at HighBeam Research.

Premium Content Partnership | HighBeam Research provides an in-depth online archive library of reference works.