A Study of Real Data.
Abstract
Nine sets of real data are analyzed. The distributions within the contaminated exponential power family which best describe these data sets are obtained by maximum likelihood. It appears that heavy tailed distributions are often produced by secular inhomogeneity in mean and variance. A specific robust estimator would behave well if, in the real world, data occurred of the kind which favored it. In this paper more classical sets of data are considered and their characteristics are studied in relation to proposed robust estimators.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 1979
- Accession Number
- ADA079737
Entities
People
- George E. P. Box
- Gina Chen
Organizations
- University of Wisconsin–Madison