Abstract
We evaluate MCMC sampling schemes for a variety of link functions in generalized linear models with Dirichlet process random effects. First, we find that there is a large amount of variability in the performance of MCMC algorithms, with the slice sampler typically being less desirable than either a Kolmogorov–Smirnov mixture representation or a Metropolis–Hastings algorithm. Second, in fitting the Dirichlet process, dealing with the precision parameter has troubled model specifications in the past. Here we find that incorporating this parameter into the MCMC sampling scheme is not only computationally feasible, but also results in a more robust set of estimates, in that they are marginalized-over rather than conditioned-upon. Applications are provided with social science problems in areas where the data can be difficult to model, and we find that the nonparametric nature of the Dirichlet process priors for the random effects leads to improved analyses with more reasonable inferences.
Similar content being viewed by others
References
Abramowitz M, Stegun IA (1972) Stirling numbers of the second kind. Section 24.1.4. In: Handbook of mathematical functions with formulas, graphs, and mathematical tables, 9th printing. Dover, New York, pp 824–825.
Albert JH, Chib S (1993) Bayesian analysis of binary and polychotomous response data. J Am Stat Assoc 88: 669–679
Andrews DF, Mallows CL (1974) Scale mixtures of normal distributions. J R Stat Soc Ser B 36: 99–102
Antoniak CE (1974) Mixtures of Dirichlet processes with applications to Bayesian nonparametric problems. Ann Stat 2: 1152–1174
Balakrishnan N (1992) Handbook of the logistic distribution. CRC Press, Boca Raton
Blackwell D, MacQueen JB (1973) Discreteness of Ferguson selections. Ann Stat 1: 358–365
Breslow NE, Clayton DG (1993) Approximate inference in generalized linear mixed models. J Am Stat Assoc 88: 9–25
Buonaccorsi JP (1996) Measurement error in the response in the general linear model. J Am Stat Assoc 91: 633–642
Chib S, Greenberg E, Chen Y (1998) MCMC methods for fitting and comparing multinomial response models. Technical Report, Economics Working Paper Archive, Washington University at St. Louis, http://129.3.20.41/econ-wp/em/papers/9802/9802001.pdf
Chib S, Winkelmann R (2001) Markov chain Monte Carlo analysis of correlated count data. J Bus Econ Stat 19: 428–435
Damien P, Wakefield J, Walker S (1999) Gibbs sampling for Bayesian non-conjugate and hierarchical models by using auxiliary variables. J R Stat Soc Ser B 61: 331–344
Devroye L (1986) Non-uniform random variate generation. Springer, New York
Dey DK, Ghosh SK, Mallick BK (2000) Generalized linear models: a Bayesian perspective. Marcel Dekker, New York
Dorazio RM, Mukherjee B, Zhang L, Ghosh M, Jelks HL, Jordan F (2007) Modelling unobserved sources of heterogeneity in animal abundance using a Dirichlet process prior. Biometrics (online publication August 3, 2007)
Escobar MD, West M (1995) Bayesian density estimation and inference using mixtures. J Am Stat Assoc 90: 577–588
Fahrmeir L, Tutz G (2001) Multivariate statistical modelling based on generalized linear models. 2. Springer, New York
Ferguson TS (1973) A Bayesian analysis of some nonparametric problems. Ann Stat 1: 209–230
Gill J, Casella G (2009) Nonparametric priors for ordinal Bayesian social science models: specification and estimation. J Am Stat Assoc 104: 453–464
Gurr TR, Marshall MG, Jaggers K (2003) PolityIV, http://www.cidcm.umd.edu/inscr/polity/
Ishwaran H, James LF (2001) Gibbs sampling methods for stick-breaking priors. J Am Stat Assoc 96: 161–173
Jiang J (2007) Linear and generalized linear mixed models and their applications. Springer, New York
Koch MT, Cranmer S (2007) Terrorism than governments of the right? Testing the ‘Dick Cheney’ hypothesis: do governments of the left attract more than governments of the right?. Conflict Manage Peace Sci 24: 311–326
Korwar RM, Hollander M (1973) Contributions to the theory of Dirichlet processes. Ann Probab 1: 705–711
Kyung M, Gill J, Casella G (2009) Characterizing the variance improvement in linear Dirichlet random effects models. Stat Probab Lett 79: 2343–2350
Kyung M, Gill J, Casella G (2010) Estimation in Dirichlet random effects models. Ann Stat 38: 979–1009
Kyung M, Gill J, Casella G (2011) New findings from terrorism data: Dirichlet process random effects models for latent groups. J R Stat Soc Ser C (Forthcoming)
Liu JS (1996) Nonparametric hierarchical Bayes via sequential imputations. Ann Stat 24: 911–930
Lo AY (1984) On a class of Bayesian nonparametric estimates: I. Density estimates. Ann Stat 12: 351–357
MacEachern SN, Müller P (1998) Estimating mixture of Dirichlet process model. J Comput Graph Stat 7: 223–238
McAuliffe JD, Blei DM, Jordan MI (2006) Nonparametric empirical Bayes for the Dirichlet process mixture model. Stat Comput 16: 5–14
McCullagh P, Nelder JA (1989) Generalized linear models. 2. Chapman & Hall, New York
McCullagh P, Yang J (2006) Stochastic classification models. Int Congr Math III: 669–686
McCulloch CE, Searle SR (2001) Generalized, linear, and mixed models. Wiley, New York
Mira A, Tierney L (2002) Efficiency and convergence properties of slice samplers. Scand J Stat 29: 1–12
Neal RM (2000) Markov chain sampling methods for Dirichlet process mixture models. J Comput Graph Stat 9: 249–265
Neal RM (2003) Slice sampling. Ann Stat 31: 705–741
Robert C, Casella G (2004) Monte Carlo statistical methods. Springer, New York
Sethuraman J (1994) A constructive definition of Dirichlet priors. Statistica Sinica 4: 639–650
Teh YW, Jordan MI, Beal MJ, Blei DM (2006) Hierarchical Dirichlet processes. J Am Stat Assoc 101: 1566–1581
Tierney L, Kadane JB (1986) Accurate approximations for posterior moments and marginal densities. J Am Stat Assoc 81: 82–86
Wang N, Lin X, Gutierrez RG, Carroll RJ (1998) Bias analysis and SIMEX approach in generalized linear mixed measurement error models. J Am Stat Assoc 93: 249–261
West M (1987) On scale mixtures of normal distributions. Biometrika 74: 646–648
Wolfinger R, O’Connell M (1993) Generalized linear mixed models: a pseudolikelihood approach. J Stat Comput Simul 48: 233–243
Author information
Authors and Affiliations
Corresponding author
Additional information
This study was supported by National Science Foundation Grants DMS-0631632, SES-0631588, DMS-04-05543.
Rights and permissions
About this article
Cite this article
Kyung, M., Gill, J. & Casella, G. Sampling schemes for generalized linear Dirichlet process random effects models. Stat Methods Appl 20, 259–290 (2011). https://doi.org/10.1007/s10260-011-0168-x
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10260-011-0168-x