An Empirical Evaluation of Five Circular Error Probable Estimation Techniques and a Method for Improving Them

Author

William L. Tongue

Date of Award

3-1993

Document Type

Thesis

Degree Name

Master of Science

Department

Department of Operational Sciences

First Advisor

Paul F. Auclair, PhD

Abstract

This study compared five CEP estimation techniques under the assumption that the crossrange and downrange miss distances of the sample data follow a bivariate normal distribution. The analysis determined the sensitivities of these models to changes in sample size, bias, correlation, and ellipticity in terms of three measures of effectiveness: Mean relative error (RE), variance of RE, and mean squared error (MSE) of RE. In general, it was found that sample size was the most significant parameter in determining the best CEP method. Mean RE provided a 'strong' distinction between estimators, while variance and MSE provided 'weak' distinctions between estimators. An attempt to improve one of the better estimation techniques using least squares regression proved quite successful.

AFIT Designator

AFIT-GST-ENS-93M-13

DTIC Accession Number

ADA266528

Comments

The author's Vita page is omitted.

Recommended Citation

Tongue, William L., "An Empirical Evaluation of Five Circular Error Probable Estimation Techniques and a Method for Improving Them" (1993). Theses and Dissertations. 7248.
https://scholar.afit.edu/etd/7248