The Information in Contingency Tables - An Application of Information- Theoretic Concepts to the Analysis of Contingency Tables

Abstract

The analysis of the information in contingency tables is an aspect of multivariate (multiple variates) analysis with particular application to qualitative or categorical as well as quantitative variables. The analysis is concerned with counts in multiway cross-classifications or multiway contingency tables. Multiway contingency tables, or cross-classifications of vectors of discrete random variables, provide a useful approach to the analysis of multivariate discrete data. The method of analysis presented will bring out the various inter-relationships among the classificatory variables in a multiway cross-classification or contingency table in many dimensions. The procedure is based on the Principle of Minimum Discrimination Information Estimation, associated statistics and Analysis of Information. General computer programs are available to provide the necessary results for inference. An analysis of a four-way contingency table is presented for illustration of these techniques.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 04, 1976
Accession Number
ADA030695

Entities

People

  • C. T. Ireland
  • S. Kullback

Organizations

  • Stanford University

Tags

Communities of Interest

  • C4I
  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Computer Programs
  • Computers
  • Confidence Limits
  • Data Science
  • Education
  • Families (Human)
  • Information Science
  • Men
  • North Carolina
  • Plastic Explosives
  • Probability
  • Probability Distributions
  • Random Variables
  • Regression Analysis
  • Statistics
  • Surveys
  • United States

Fields of Study

  • Mathematics

Readers

  • Regression Analysis.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference