Composite likelihood inference by nonparametric saddlepoint tests

The class of composite likelihood functions provides a flexible and powerful toolkit to carry out approximate inference for complex statistical models when the full likelihood is either impossible to specify or unfeasible to compute. However, the strength of the composite likelihood approach is dimmed when considering hypothesis testing about a multidimensional parameter because the finite sample behavior of likelihood ratio, Wald, and score-type test statistics is tied to the Godambe information matrix. Consequently, inaccurate estimates of the Godambe information translate in inaccurate p-values. The approach based on a fully nonparametric saddlepoint test statistic derived from the composite score functions is shown to achieve accurate inference. The proposed statistic is asymptotically chi-squared distributed up to a relative error of second order and does not depend on the Godambe information. The validity of the method is demonstrated through simulation studies.

[1]  B. Ripley,et al.  Robust Statistics , 2018, Encyclopedia of Mathematical Geosciences.

[2]  Wenxin Jiang,et al.  The Indirect Method: Inference Based on Intermediate Statistics—A Synthesis and Examples , 2004 .

[3]  Yanyuan Ma,et al.  Saddlepoint Test in Measurement Error Models , 2011 .

[4]  Patrick J. Heagerty,et al.  Weighted empirical adaptive variance estimators for correlated data regression , 1999 .

[5]  P. Donnelly,et al.  The Fine-Scale Structure of Recombination Rate Variation in the Human Genome , 2004, Science.

[6]  G. Molenberghs,et al.  Models for Discrete Longitudinal Data , 2005 .

[7]  D. Cox,et al.  A note on pseudolikelihood constructed from marginal densities , 2004 .

[8]  Richard E. Chandler,et al.  Inference for clustered data using the independence loglikelihood , 2007 .

[9]  Ruggero Bellio,et al.  A pairwise likelihood approach to generalized linear models with crossed random effects , 2005 .

[10]  E. Ronchetti,et al.  Robust Bounded-Influence Tests in General Parametric Models , 1994 .

[11]  Elvezio Ronchetti,et al.  Robust small sample accurate inference in moment condition models , 2012, Comput. Stat. Data Anal..

[12]  N. Reid,et al.  AN OVERVIEW OF COMPOSITE LIKELIHOOD METHODS , 2011 .

[13]  H. Daniels.  Saddlepoint Approximations in Statistics , 1954 .

[14]  Zhiyi Chi,et al.  Approximating likelihoods for large spatial data sets , 2004 .

[15]  D. Zimmerman,et al.  Towards reconciling two asymptotic frameworks in spatial statistics , 2005 .

[16]  Harry Joe,et al.  Composite Likelihood Methods , 2012 .

[17]  R. Hudson.  Two-locus sampling distributions and their application. , 2001, Genetics.

[18]  Elvezio Ronchetti,et al.  Saddlepoint approximations and tests based on multivariate M-estimates , 2003 .

[19]  K. Heggland,et al.  Estimating functions in indirect inference , 2004 .

[20]  S. Lele,et al.  A Composite Likelihood Approach to Binary Spatial Data , 1998 .

[21]  R. Butler.  SADDLEPOINT APPROXIMATIONS WITH APPLICATIONS. , 2007 .

[22]  Xiaotong Shen,et al.  Empirical Likelihood , 2002 .

[23]  Øivind Skare,et al.  Pairwise likelihood inference in spatial generalized linear mixed models , 2005, Comput. Stat. Data Anal..

[24]  Anthony C. Davison,et al.  Applied Asymptotics: Case Studies in Small-Sample Statistics , 2007 .

[25]  N. Jewell,et al.  Hypothesis testing of regression parameters in semiparametric generalized linear models for cluster correlated data , 1990 .

[26]  S. Padoan,et al.  Likelihood-Based Inference for Max-Stable Processes , 2009, 0902.3060.

[27]  Gerda Claeskens,et al.  Bootstrap tests for misspecified models, with application to clustered binary data , 2001 .

[28]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[29]  P. Moral,et al.  Sequential Monte Carlo samplers , 2002, cond-mat/0212648.

[30]  Luigi Pace,et al.  ADJUSTING COMPOSITE LIKELIHOOD RATIO STATISTICS , 2009 .

[31]  Bruce G. Lindsay,et al.  ISSUES AND STRATEGIES IN THE SELECTION OF COMPOSITE LIKELIHOODS , 2011 .

[32]  Geert Molenberghs,et al.  A pairwise likelihood approach to estimation in multilevel probit models , 2004, Comput. Stat. Data Anal..

[33]  Saddlepoint approximations for multivariate M-estimates with applications to bootstrap accuracy , 2008 .

[34]  Geert Verbeke,et al.  Pairwise Fitting of Mixed Models for the Joint Modeling of Multivariate Longitudinal Profiles , 2006, Biometrics.

[35]  Laura Ventura,et al.  Bayesian composite marginal likelihoods , 2011 .

[36]  J. Kent.  Robust properties of likelihood ratio tests , 1982 .

[37]  Kuldeep Kumar.  Robust Statistics, 2nd edition by P.J. Huber & E.M. Ronchetti [book review] , 2011 .