Using Biweights in the Two-Sample Problem.
Abstract
It is proposed to replace the usual Student's-t statistic which tests for equality of means of two distributions and to construct a confidence interval for the difference by a biweight-'t' statistic. The biweight-'t' is a ratio of the difference of the biweight estimates of location from the two samples to a standard error of this difference. Three different forms of the denominator are evaluated. Monte Carlo simulations reveal that resulting confidence intervals are highly efficient in moderate sample sizes, and that nominal levels are nearly attained, even when considering extreme percentage points.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 1979
- Accession Number
- ADA084844
Entities
People
- Karen Kafadar
Organizations
- Princeton University