Selection of Variables in Discriminant Analysis.

Abstract

In a number of disciplines, data analysts are confronted with the problem of classifying an observation into one of the distinct groups when the number of variables is very large. So, it is of interest to find out a smaller number of important variables which are adequate for discrimination. These variables may be a subset of the original variables or certain linear combinations of the original variables. The selection of variables is important since there are situations where inclusion of unimportant variables may actually decrease the ability for discrimination. Apart from it, it is more feasible to analyze the data from cost and computational considerations if the number of variables is small. In this chapter, we discuss various procedures for the selection of variables in discriminant analysis. In Section 2, we discuss procedures to find out whether certain discriminant coefficients associated with variables are important for discrimination between two populations. In Section 3, generalizations of the above procedures for several populations are discussed. In Section 4, we discuss various procedures to determine the number of important discriminant functions.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 1982
Accession Number
ADA121143

Entities

People

  • Paruchuri R. Krishnaiah

Organizations

  • University of Pittsburgh

Tags

DTIC Thesaurus Topics

  • Air Force
  • Coefficients
  • Correlation Analysis
  • Covariance
  • Data Science
  • Discriminant Analysis
  • Discrimination
  • Eigenvalues
  • Hypotheses
  • Information Science
  • Multivariate Analysis
  • New York
  • Observation
  • Scientific Research
  • Statistics
  • United States
  • Wishart Matrices

Fields of Study

  • Mathematics

Readers

  • Business Analytics
  • Regression Analysis.