A Simple Method for Robust Regression.
Abstract
Estimates of the parameters of a linear model are usually obtained by the method of ordinary least-squares (OLS), which is sensitive to large values of the additive error term. By dividing the sample into non-overlapping subsamples and computing the trimmed means of OLS subsample regression coefficients, the authors obtain a simple consistent and asymptotically normal initial estimate of the coefficients which can be used on one of the various robust techniques which have been recently discussed in the literature, or which can be used to trim the sample observations which have large residuals. (Modified author abstract)
Document Details
- Document Type
- Technical Report
- Publication Date
- May 07, 1974
- Accession Number
- AD0781474
Entities
People
- Melvin J. Hinich
- Prem P. Talwar
Organizations
- Carnegie Mellon University