Information Analysis of Linear Interactions in Contingency Tables

Abstract

The use of the minimum discrimination information (MDI) approach in studying null hypotheses of no interactions on a linear scale in contingency tables of one response many factors type is illustrated. Quadratic approximations to the M.D.I. statistic are related to Wald-type statistics and Neyman's modified chi-square. Follow up analyses when the null hypothesis is not satisfied are illustrated.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 15, 1977
Accession Number
ADA044902

Entities

People

  • D. V. Gokhale
  • S. Kullback

Organizations

  • Stanford University

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Accidents
  • Algorithms
  • Classification
  • Data Science
  • Discrimination
  • Frequency
  • Hypotheses
  • Information Science
  • Military Research
  • New York
  • Numbers
  • Probability
  • Probability Distributions
  • Security
  • Statistics
  • Three Dimensional
  • United States

Fields of Study

  • Mathematics

Readers

  • Regression Analysis.
  • Statistical inference.