Bayesian Inference with Missing Data Using Bound and Collapse

Abstract Current Bayesian methods to estimate conditional probabilities from samples with missing data pose serious problems of robustness and computational efficiency. This article introduces a new method, called bound and collapse (BC) to tackle these problems. BC first bounds the possible estimates consistent with the available information and then collapses these bounds to a point estimate using information about the pattern of missing data. Deterministic approximations of the variance and of the posterior distribution are proposed, and their accuracy is compared to stochastic approximations in a real dataset of polling data subject to nonresponse.

[1]  A. F. Smith,et al.  Statistical analysis of finite mixture distributions , 1986 .

[2]  S. Brodetsky.  Essai philosophique sur les probabilités , 1922, Nature.

[3]  David Butler,et al.  The British general election of 1983 , 1984 .

[4]  I. Good,et al.  The Estimation of Probabilities: An Essay on Modern Bayesian Methods. , 1967 .

[5]  J. Copas,et al.  Inference for Non‐random Samples , 1997 .

[6]  L. Wasserman,et al.  Inferences from multinomial data: Learning about a bag of marbles - Discussion , 1996 .

[7]  Donald Geman,et al.  Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8]  D. Rubin,et al.  Handling “Don't Know” Survey Responses: The Case of the Slovenian Plebiscite , 1995 .

[9]  Joseph B. Kadane,et al.  Bayesian Methods for Censored Categorical Data , 1987 .

[10]  Henry E. Kyburg,et al.  Probability and the logic of rational belief , 1970 .

[11]  S. Kotz,et al.  Symmetric Multivariate and Related Distributions , 1989 .

[12]  D. Rubin,et al.  Statistical Analysis with Missing Data. , 1989 .

[13]  David E. Booth,et al.  Analysis of Incomplete Multivariate Data , 2000, Technometrics.

[14]  J. Berger.  Statistical Decision Theory and Bayesian Analysis , 1988 .

[15]  Anton Meredith Brink.  Bayesian analysis of contingency tables , 1997 .

[16]  P. Walley.  Statistical Reasoning with Imprecise Probabilities , 1990 .

[17]  Joseph B. Kadane,et al.  Subjective Bayesian analysis for surveys with missing data , 1993 .

[18]  R. F.,et al.  Mathematical Statistics , 1944, Nature.

[19]  W. Wong,et al.  The calculation of posterior distributions by data augmentation , 1987 .

[20]  Jonathan J. Forster,et al.  Monte Carlo exact conditional tests for log-linear and logistic models , 1996 .

[21]  Sheldon M. Ross,et al.  Stochastic Processes , 2018, Gauge Integral Structures for Stochastic Calculus and Quantum Electrodynamics.

[22]  D. Rubin.  INFERENCE AND MISSING DATA , 1975 .

[23]  Irving John Good,et al.  Subjective Probability as the Measure of a Non-measurable Set , 1962 .

[24]  Carlos Alberto de Bragança Pereira,et al.  Bayesian methods for categorical data under informative general censoring , 1995 .

[25]  R. Worcester.  The British General Election of 1992 , 1993 .

[26]  Joseph B. Kadane,et al.  Missing Data in the Forensic Context , 1997 .

[27]  Andrew L. Rukhin,et al.  Tools for statistical inference , 1991 .

[28]  P. Walley.  Inferences from Multinomial Data: Learning About a Bag of Marbles , 1996 .

[29]  David W. Scott.  The New S Language , 1990 .

[30]  Walter R. Gilks,et al.  Strategies for improving MCMC , 1995 .

[31]  Ross Ihaka,et al.  Gentleman R: R: A language for data analysis and graphics , 1996 .

[32]  J. Forster,et al.  Monte Carlo exact tests for square contingency tables , 1996 .

[33]  Jonathan J. Forster,et al.  Model‐based inference for categorical survey data subject to non‐ignorable non‐response , 1998 .

[34]  Thomas J. Jiang,et al.  Computation of Carlson's Multiple Hypergeometric Function R for Bayesian Applications , 1992 .

[35]  David J. Spiegelhalter,et al.  Sequential updating of conditional probabilities on directed graphical structures , 1990, Networks.

[36]  A. Dawid,et al.  A comparison of sequential learning methods for incomplete data , 1995 .

[37]  Toshinobu Kashiwabara,et al.  Efficient algorithms for finding maximum cliques of an overlap graph , 1990, Networks.

[38]  Taesung Park,et al.  Models for Categorical Data with Nonignorable Nonresponse , 1994 .

[39]  C. Pereira,et al.  Bayesian analysis of categorical data informatively censored , 1992 .