Convergence of Dirichlet Measures and the Interpretation of Their Parameter.

作者: Jayaram Sethuraman , Ram C. Tiwari

DOI: 10.1016/B978-0-12-307502-4.50023-4

关键词:

摘要: The form of the Bayes estimate population mean with respect to a Dirichlet prior parameter has given rise interpretation that ( X ) is sample size. Furthermore, if made tend zero, then mathematically converges classical estimator, is, mean. This further general feeling allowing become small not only makes size but also it corresponds no information. By investigating limits distributions as tends various values, misleading think and smallness In fact, very values actually lot information concerning unknown true distribution would be generally unacceptable statistician.

参考文章(6)
Hajime Yamato, RELATIONS BETWEEN LIMITING BAYES ESTIMATES AND U-STATISTICS FOR ESTIMABLE PARAMETERS Journal of the Japan Statistical Society. Japanese issue. ,vol. 7, pp. 57- 66 ,(1977) , 10.11329/JJSS1970.7.57
David Blackwell, Discreteness of Ferguson Selections Annals of Statistics. ,vol. 1, pp. 356- 358 ,(1973) , 10.1214/AOS/1176342373
Hajime Yamato, Relations between limiting bayes estimates and the u-statistics for estimable parameters of degrees 2 and 3 Communications in Statistics-theory and Methods. ,vol. 6, pp. 55- 66 ,(1977) , 10.1080/03610927708827469
Yu. V. Prokhorov, Convergence of Random Processes and Limit Theorems in Probability Theory Theory of Probability & Its Applications. ,vol. 1, pp. 157- 214 ,(1956) , 10.1137/1101016
Thomas S. Ferguson, A Bayesian Analysis of Some Nonparametric Problems Annals of Statistics. ,vol. 1, pp. 209- 230 ,(1973) , 10.1214/AOS/1176342360
David Blackwell, James B. MacQueen, Ferguson Distributions Via Polya Urn Schemes Annals of Statistics. ,vol. 1, pp. 353- 355 ,(1973) , 10.1214/AOS/1176342372