Abstract

The effect of the median polish procedure on artificial data having known general characteristics, disturbances, and random elements is examined. Monte Carlo experiments were designed for this purpose. The behavior of the fits, residuals, and various accompanying statistics are determined under special controlled conditions. Then, the procedure is applied to selected data sets from the literature utilizing the findings of this study.

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Publication Date

8-1982

Comments

Originally presented at the Social Statistics Section of the American Statistical Association, 1982 Annual Meeting.

Document Type

Conference Paper

Department, Program, or Center

Mathematics and Statistics, School of

College

College of Science

Campus

RIT – Main Campus

Share

COinS