DocumentCode
2384321
Title
The consistency of the BIC Markov order estimator
Author
Csiszar, Imre ; Shields, Paul C.
Author_Institution
Inst. of Math., Hungarian Acad. of Sci., Budapest, Hungary
fYear
2000
fDate
2000
Firstpage
26
Abstract
We show that the BIC (Bayesian information criterion) estimator of the order of a Markov chain (with finite alphabet) gives the correct order, eventually almost surely as the sample size goes to ∞, thereby strengthening earlier consistency results that assumed an apriori bound on the order. A key tool is a strong typicality result for Markov sample paths. We also show that the Bayesian or MDL estimator, of which the BIC estimator is regarded as an approximation, fails to be consistent for the uniformly distributed i.i.d. process
Keywords
Bayes methods; Markov processes; estimation theory; information theory; BIC Markov order estimator; Bayesian information criterion; MDL estimator; Markov chain; Markov sample paths; consistency; independent identically distributed process; uniformly distributed i.i.d. process; Bayesian methods; Estimation theory; Mathematics; Statistics;
fLanguage
English
Publisher
ieee
Conference_Titel
Information Theory, 2000. Proceedings. IEEE International Symposium on
Conference_Location
Sorrento
Print_ISBN
0-7803-5857-0
Type
conf
DOI
10.1109/ISIT.2000.866316
Filename
866316
Link To Document