Rate of convergence of predictive distributions for dependent data
Berti, Patrizia ; Crimaldi, Irene ; Pratelli, Luca ; Rigo, Pietro
Bernoulli, Tome 15 (2009) no. 1, p. 1351-1367 / Harvested from Project Euclid
This paper deals with empirical processes of the type ¶ \[C_{n}(B)=\sqrt{n}\{\mu_{n}(B)-P(X_{n+1}\in B\mid X_{1},\ldots,X_{n})\},\] ¶ where (Xn) is a sequence of random variables and μn=(1/n)∑i=1nδXi the empirical measure. Conditions for supB|Cn(B)| to converge stably (in particular, in distribution) are given, where B ranges over a suitable class of measurable sets. These conditions apply when (Xn) is exchangeable or, more generally, conditionally identically distributed (in the sense of Berti et al. [Ann. Probab. 32 (2004) 2029–2052]). By such conditions, in some relevant situations, one obtains that $\sup_{B}|C_{n}(B)|\stackrel{P}{\rightarrow}0$ or even that $\sqrt{n}\sup_{B}|C_{n}(B)|$ converges a.s. Results of this type are useful in Bayesian statistics.
Publié le : 2009-11-15
Classification:  Bayesian predictive inference,  central limit theorem,  conditional identity in distribution,  empirical distribution,  exchangeability,  predictive distribution,  stable convergence
@article{1262962239,
     author = {Berti, Patrizia and Crimaldi, Irene and Pratelli, Luca and Rigo, Pietro},
     title = {Rate of convergence of predictive distributions for dependent data},
     journal = {Bernoulli},
     volume = {15},
     number = {1},
     year = {2009},
     pages = { 1351-1367},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1262962239}
}
Berti, Patrizia; Crimaldi, Irene; Pratelli, Luca; Rigo, Pietro. Rate of convergence of predictive distributions for dependent data. Bernoulli, Tome 15 (2009) no. 1, pp.  1351-1367. http://gdmltest.u-ga.fr/item/1262962239/