Sufficiency and partitions of the class of all possible discrete distributions

Glen Meeden

Research output: Contribution to journalArticlepeer-review

1 Scopus citations


For discrete statistical models it is shown that any statistic induces a partition of the set of all possible distributions defined on the sample space. This partition identifies the subsets of the parameter space for which the statistic is sufficient.

Original languageEnglish (US)
Pages (from-to)42-44
Number of pages3
JournalAmerican Statistician
Issue number1
StatePublished - Feb 1986

Bibliographical note

Funding Information:
*Glen Meeden is Professor, Department of Statistics, Iowa State University, Ames, IA 5001 1. Research was supported by National Science Foundation Grant MCS-8401740.


  • Discrete probability
  • Sufficient statistics


Dive into the research topics of 'Sufficiency and partitions of the class of all possible discrete distributions'. Together they form a unique fingerprint.

Cite this