Accession Number : ADA305773

Title :   A Hypothesis-Testing Approach to Outlier Detection.

Descriptive Note : Final rept. Sep 93-Sep 95,

Corporate Author : SOUTHERN METHODIST UNIV DALLAS TX DEPT OF STATISTICAL SCIENCE

Personal Author(s) : Woodward, W. A. ; Gray, H. L.

PDF Url : ADA305773

Report Date : OCT 1995

Pagination or Media Count : 128

Abstract : In this report we develop a general robust near-optimal methodology for testing for outliers to one or more populations. The methodology requires no distributional assumptions and allows imperfect data sets, i.e., data sets with missing observations. The data can be from one or several seismic stations or can be a mixture of seismic and nonseismic data and continuous and categorical data. Listings of the computer code required to implement the methodology are included. (AN)

Descriptors :   *MAXIMUM LIKELIHOOD ESTIMATION, *HYPOTHESES, *DISCRIMINATE ANALYSIS, DATA BASES, COMPUTER PROGRAM DOCUMENTATION, MATHEMATICAL MODELS, ALGORITHMS, COMPUTERIZED SIMULATION, NUCLEAR PROLIFERATION, OPTIMIZATION, PARAMETRIC ANALYSIS, MONITORING, MULTIVARIATE ANALYSIS, STATISTICAL TESTS, RANDOM VARIABLES, STATISTICAL DATA, NUCLEAR EXPLOSION TESTING, NONPARAMETRIC STATISTICS, SEISMIC DATA, APPROXIMATION(MATHEMATICS), ERROR ANALYSIS, DATA COMPRESSION, COVARIANCE, SEISMIC DISCRIMINATION, EARTHQUAKES, APPLIED MATHEMATICS, BLAST, NORMAL DISTRIBUTION, SEISMOLOGICAL STATIONS, POPULATION(MATHEMATICS).

Subject Categories : Statistics and Probability
      Seismology
      Operations Research

Distribution Statement : APPROVED FOR PUBLIC RELEASE