scientific article; zbMATH DE number 1339024

MaRDI QIDQ4264378zbMATH OpenFDO

Authors Persi Diaconis

Publication date 22 September 1999

zbMATH Keywords

exchangeabilityGibbs statesgroup invariancestatistical mechanicssufficiency

Mathematics Subject Classification ID

Exchangeability for stochastic processes (60G09) Discrete-time Markov processes on general state spaces (60J05) Foundations and philosophical topics in statistics (62A01) Sufficient statistics and fields (62B05)

Review text

Sufficiency is a theoretical tool that has grown up in mathematical statistics, and may be described crudely as the theory of how much data can be thrown away. The present paper reviews the basic achievements of the theory in statistical problems and sketches applications in other areas of mathematics. The author shows how the idea of sufficiency gives a suitable framework for exchangeability (an important piece of the Bayesian theory of statistics) and Gibbs states (the rigorous theory of phase transitions in statistical mechanics). In these two settings, sufficiency may be seen as a sweeping generalization of group invariance. The paper is organized as follows: Section 2 reviews the history and the main mathematical results of sufficiency. Section 3 introduces exchangeability as part of the Bayesian view of statistics. Section 4 shows how sufficiency ideas give a natural foundation for exchangeability. For the entire collection see [Zbl 0921.00016].

Recommendations

• Sufficiency in bayesian models
• scientific article; zbMATH DE number 1064831
• On conditional independence and the relationship between sufficiency and invariance under the Bayesian point of view

Cited in

(10)

• Self-similarity of mathematical likelihood
• scientific article; zbMATH DE number 3858151 (Why is no real title available?)
• scientific article; zbMATH DE number 1064831 (Why is no real title available?)
• Sufficiency in bayesian models
• scientific article; zbMATH DE number 218954 (Why is no real title available?)
• Probabilistic symmetries and invariant neural networks
• Tropical sufficient statistics for persistent homology
• Emerging directions in Bayesian computation
• Statistics for phylogenetic trees
• Perfect aggregation of Bayesian analysis on compositional data

This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4264378)