Bias Reduction When There Is No Unbiased Estimate.
Abstract
Let phi be a parameter for which there is no unbiased estimator. This note shows that for an arbitrary sequence of estimators T superscript(k), if the bias of T superscript(k) tend to O then their variances must tend to infinity. Keywords: Bias reduction; Jackknife estimate of bias; Bootstrap estimate of bias.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1988
- Accession Number
- ADA189407
Entities
People
- Hani Doss
- Jayaram Sethuraman
Organizations
- Florida State University