A BAYESIAN APPROACH TO SOME MISSING VALUE PROBLEMS IN ANOVA AND CONTINGENCY TABLES
JOHNS HOPKINS UNIV BALTIMORE MD DEPT OF STATISTICS
Pagination or Media Count:
Non-Bayesian procedures for dealing with missing values in ANOVA and contingency tables are well-known. This paper derives the appropriate Bayesian procedures for a randomized block design and for and r x c contingency table. The posterior distributions for the missing observation are given. This paper shows that the classical non-Bayesian procedures do have a simple and natural Bayesian interpretation.
- Statistics and Probability