×

Nonasymptotic approach to Bayesian semiparametric inference. (English. Russian original) Zbl 1353.62034

Dokl. Math. 93, No. 2, 155-158 (2016); translation from Dokl. Akad. Nauk, Ross. Akad. Nauk 467, No. 2, 139-143 (2016).
Summary: The classical semiparametric Bernstein-von Mises (BvM) results is reconsidered in a non-classical setup allowing finite samples and model misspecication. We obtain an upper bound on the error of Gaussian approximation of the posterior distribution for the target parameter which is explicit in the dimension of the target parameter and in the dimension of sieve approximation of the nuisance parameter. This helps to identify the so called critical dimension \(p_n\) of the sieve approximation of the full parameter for which the BvM result is applicable. If the bias induced by sieve approximation is small and dimension of sieve approximation is smaller then critical dimension than the BvM result is valid. In the important i.i.d. and regression cases, we show that the condition “\(p_n^2q/n\) is small”, where \(q\) is the dimension of the target parameter and \(n\) is the sample size, leads to the BvM result under general assumptions on the model.

MSC:

62G05 Nonparametric estimation
62F15 Bayesian inference
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] L. Le Cam and G. L. Yang, Asymptotics in Statistics: Some Basic Concepts (Springer, Berlin, 1990). · Zbl 0719.62003 · doi:10.1007/978-1-4684-0377-0
[2] P. Buhlmann and S. van de Geer, Statistics for HighDimensional Data: Methods, Theory, and Applications (Springer, Berlin, 2011). · Zbl 1273.62015 · doi:10.1007/978-3-642-20192-9
[3] Schwartz, L., No article title, Probab. Theory Relat. Fields, 4, 10-26 (1965) · Zbl 0158.17606
[4] Barron, A.; Schervish, M. J.; Wasserman, L., No article title, Ann. Stat., 27, 536-561 (1996) · Zbl 0980.62039
[5] Shen, X., No article title, J. Am. Stat. Assoc., 97, 222-235 (2002) · Zbl 1073.62517 · doi:10.1198/016214502753479365
[6] Cheng, G.; Kosorok, M. R., No article title, Ann. Stat., 36, 1819-1853 (2008) · Zbl 1142.62031 · doi:10.1214/07-AOS536
[7] Bickel, P. J.; Kleijn, B. J. K., No article title, Ann. Stat., 40, 206-237 (2012) · Zbl 1246.62081 · doi:10.1214/11-AOS921
[8] Castillo, I., No article title, Probab. Theory Relat.Fields, 152, 53-99 (2012) · Zbl 1232.62054 · doi:10.1007/s00440-010-0316-5
[9] Rivoirard, V.; Rousseau, J., No article title, Ann. Stat., 40, 1489-1523 (2012) · Zbl 1257.62036 · doi:10.1214/12-AOS1004
[10] I. Castillo and J. Rousseau, Available at arXiv:1305.4482 [math.ST].
[11] Chentsov, N. N., No article title, Dokl. Akad. Nauk SSSR, 147, 45-48 (1962)
[12] Panov, M.; Spokoiny, V., No article title, Bayesian Anal., 10, 665-710 (2015) · Zbl 1335.62057 · doi:10.1214/14-BA926
[13] Zaytsev, A.; Burnaev, E.; Spokoiny, V., No article title, J. Math. Sci., 203, 789-798 (2014) · Zbl 1310.62092 · doi:10.1007/s10958-014-2168-5
[14] Andresen, A.; Spokoiny, V., No article title, Electron. J. Stat., 8, 3077-3125 (2014) · Zbl 1308.62031 · doi:10.1214/14-EJS982
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.