Posterior propriety and admissibiity of hyperpriors in normal hierarchical models. (English) Zbl 1068.62005

Summary: Hierarchical modeling is wonderful and here to stay, but hyperparameter priors are often chosen in a casual fashion. Unfortunately, as the number of hyperparameters grows, the effects of casual choices can multiply, leading to considerably inferior performance. As an extreme, but not uncommon, example use of the wrong hyperparameter priors can even lead to impropriety of the posterior.
For exchangeable hierarchical multivariate normal models, we first determine when a standard class of hierarchical priors results in proper or improper posteriors. We next determine which elements of this class lead to admissible estimators of the mean under quadratic loss; such considerations provide one useful guideline for choice among hierarchical priors. Finally, computational issues with the resulting posterior distributions are addressed.


62C15 Admissibility in statistical decision theory
62H10 Multivariate distribution of statistics
62H12 Estimation in multivariate analysis
62F15 Bayesian inference
Full Text: DOI arXiv


[1] Angers, J.-F. (1992). Use of the Student-\(t\) prior for the estimation of normal means: A computational approach. In Bayesian Statistics 4 (J. M. Bernardo, J. O. Berger, A. P. Dawid and A. F. M. Smith, eds.) 567–575. Oxford Univ. Press, New York.
[2] Angers, J.-F., MacGibbon, B. and Wang, S. (1998). Bayesian estimation of intra-block exchangeable normal means with applications. Sankhyā Ser. A 60 198–213. · Zbl 0976.62050
[3] Bayarri, M. J. and Berger, J. O. (2004). The interplay of Bayesian and frequentist analysis. Statist. Sci. 19 58–80. · Zbl 1062.62001
[4] Berger, J. O. (1980). A robust generalized Bayes estimator and confidence region for a multivariate normal mean. Ann. Statist. 8 716–761. JSTOR: · Zbl 0464.62026
[5] Berger, J. O. (1985). Statistical Decision Theory and Bayesian Analysis , 2nd ed. Springer, New York. · Zbl 0572.62008
[6] Berger, J. O. and Robert, C. (1990). Subjective hierarchical Bayes estimation of a multivariate normal mean: On the frequentist interface. Ann. Statist. 18 617–651. JSTOR: · Zbl 0719.62043
[7] Berger, J. O. and Strawderman, W. (1996). Choice of hierarchical priors: Admissibility in estimation of normal means. Ann. Statist. 24 931–951. · Zbl 0865.62004
[8] Bernardo, J. M. and Smith, A. F. M. (1994). Bayesian Theory . Wiley, New York. · Zbl 0796.62002
[9] Brown, L. D. (1971). Admissible estimators, recurrent diffusions, and insoluble boundary value problems. Ann. Math. Statist. 42 855–903. · Zbl 0246.62016
[10] Brown, L. D. (2000). An essay on statistical decision theory. J. Amer. Statist. Assoc. 95 1277–1281. · Zbl 1009.62502
[11] Carlin, B. P. and Louis, T. A. (2000). Bayes and Empirical Bayes Methods for Data Analysis , 2nd ed. Chapman and Hall, Boca Raton, FL. · Zbl 0871.62012
[12] Chen, M.-H., Shao, Q.-M. and Ibrahim, J. G. (2000). Monte Carlo Methods in Bayesian Computation . Springer, New York. · Zbl 0949.65005
[13] Daniels, M. (1998). Computing posterior distributions for covariance matrices. In Computing Science and Statistics. Proc. 30th Symposium on the Interface 192–196. Interface, Fairfax Station, VA.
[14] Daniels, M. and Kass, R. (1999). Nonconjugate Bayesian estimation of covariance matrices and its use in hierarchical models. J. Amer. Statist. Assoc. 94 1254–1263. · Zbl 1069.62508
[15] Daniels, M. and Pourahmadi, M. (2002). Bayesian analysis of covariance matrices and dynamic models for longitudinal data. Biometrika 89 553–566. · Zbl 1036.62019
[16] Efron, B. and Morris, C. (1972). Empirical Bayes on vector observations—an extension of Stein’s method. Biometrika 59 335–347. · Zbl 0238.62072
[17] Efron, B. and Morris, C. (1976). Multivariate empirical Bayes and estimation of covariance matrices. Ann. Statist. 4 22–32. JSTOR: · Zbl 0322.62041
[18] Everson, P. J. and Morris, C. (2000). Inference for multivariate normal hierarchical models. J. R. Stat. Soc. Ser. B Stat. Methodol. 62 399–412. · Zbl 04558580
[19] Farrell, R. H. (1985). Multivariate Calculation . Use of the Continuous Groups . Springer, New York. · Zbl 0561.62048
[20] Fourdrinier, D., Strawderman, W. and Wells, M. (1998). On the construction of Bayes minimax estimators. Ann. Statist. 26 660–671. · Zbl 0929.62004
[21] Ghosh, M. (1992). Hierarchical and empirical Bayes multivariate estimation. In Current Issues in Statistical Inference (M. Ghosh and P. K. Pathak, eds.) 151–177. IMS, Hayward, CA. · Zbl 0798.62014
[22] Haff, L. R. (1991). The variational form of certain Bayes estimators. Ann. Statist. 19 1163–1190. JSTOR: · Zbl 0739.62046
[23] Hobert, J. (2000). Hierarchical models: A current computational perspective. J. Amer. Statist. Assoc. 95 1312–1316. · Zbl 1072.62504
[24] Kotz, S., Balakrishnan, N. and Johnson, N. (2000). Continuous Multivariate Distributions 1 . Models and Applications , 2nd ed. Wiley, New York. · Zbl 0946.62001
[25] Liechty, J. C., Liechty, M. W. and Müller, P. (2004). Bayesian correlation estimation. Biometrika 91 1–14. · Zbl 1132.62314
[26] Lin, S. P. and Perlman, M. D. (1985). A Monte Carlo comparison of four estimators of a covariance matrix. In Multivariate Analysis VI (P. R. Krishnaiah, ed.) 411–429. North-Holland, Amsterdam. · Zbl 0593.62051
[27] Loh, W.-L. (1991). Estimating covariance matrices. II. J. Multivariate Anal. 36 163–174.
[28] Ni, S. and Sun, D. (2005). Bayesian estimate for vector autoregressive models. J. Bus. Econom. Statist. 23 105–117.
[29] O’Hagan, A. (1990). On outliers and credence for location parameter inference. J. Amer. Statist. Assoc. 85 172–176. · Zbl 0706.62030
[30] Perron, F. (1992). Minimax estimators of a covariance matrix. J. Multivariate Anal. 43 16–28. · Zbl 0754.62038
[31] Robert, C. P. (2001). The Bayesian Choice , 2nd ed. Springer, New York. · Zbl 0980.62005
[32] Robert, C. P. and Casella, G. (1999). Monte Carlo Statistical Methods. Springer, New York. · Zbl 0935.62005
[33] Stein, C. (1956). Inadmissibility of the usual estimator for the mean of a multivariate normal distribution. Proc. Third Berkeley Symp. Math. Statist. Probab. 1 197–206. Univ. California Press, Berkeley. · Zbl 0073.35602
[34] Stein, C. (1975). Estimation of a covariance matrix. Rietz Lecture, 39th Annual Meeting of the IMS. Atlanta, Georgia. · Zbl 0302.05015
[35] Strawderman, W. (1971). Proper Bayes minimax estimators of the multivariate normal mean. Ann. Math. Statist. 42 385–388. · Zbl 0222.62006
[36] Strawderman, W. (2000). Minimaxity. J. Amer. Statist. Assoc. 95 1364–1368. · Zbl 1072.62513
[37] Sugiura, N. and Ishibayashi, H. (1997). Reference prior Bayes estimator for bivariate normal covariance matrix with risk comparison. Comm. Statist. Theory Methods 26 2203–2221. · Zbl 0954.62539
[38] Sun, D., Tsutakawa, R. K. and He, Z. (2001). Propriety of posteriors with improper priors in hierarchical linear mixed models. Statist. Sinica 11 77–95. · Zbl 1057.62525
[39] Tang, D. (2001). Choice of priors for hierarchical models: Admissibility and computation. Ph.D. dissertation, Purdue Univ.
[40] Yang, R. and Berger, J. O. (1994). Estimation of a covariance matrix using the reference prior. Ann. Statist. 22 1195–1211. JSTOR: · Zbl 0819.62013
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. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.