Markov chain Monte Carlo test of toric homogeneous Markov chains

Markov chain models are used in various fields, such behavioral sciences or econometrics. Although the goodness of fit of the model is usually assessed by large sample approximation, it is desirable to use conditional tests if the sample size is not large. We study Markov bases for performing conditional tests of the toric homogeneous Markov chain model, which is the envelope exponential family for the usual homogeneous Markov chain model. We give a complete description of a Markov basis for the following cases: i) two-state, arbitrary length, ii) arbitrary finite state space and length of three. The general case remains to be a conjecture. We also present a numerical example of conditional tests based on our Markov basis.

[1]  Akimichi Takemura,et al.  A Markov basis for two-state toric homogeneous Markov chain model without initial parameters , 2010, 1005.1717.

[2]  Bernd Sturmfels,et al.  Higher Lawrence configurations , 2003, J. Comb. Theory, Ser. A.

[3]  Satoshi Aoki,et al.  Running Markov Chain Without Markov Bases , 2011, 1109.0078.

[4]  Giovanni Pistone,et al.  The algebra of reversible Markov chains , 2010, 1007.4282.

[5]  Sylvia Frühwirth-Schnatter,et al.  Finite Mixture and Markov Switching Models , 2006 .

[6]  Ruriko Yoshida,et al.  Degree Bounds for a Minimal Markov Basis for the Three-State Toric Homogeneous Markov Chain Model , 2011, 1108.0481.

[7]  Alan Agresti,et al.  Categorical Data Analysis , 2003 .

[8]  P. Billingsley,et al.  Statistical Methods in Markov Chains , 1961 .

[9]  Satoshi Aoki,et al.  Distance-reducing Markov bases for sampling from a discrete sample space , 2005 .

[10]  J. Ware,et al.  Issues in the analysis of repeated categorical outcomes. , 1988, Statistics in medicine.

[11]  P. Diaconis,et al.  Algebraic algorithms for sampling from conditional distributions , 1998 .

[12]  L. Pachter,et al.  Algebraic Statistics for Computational Biology: Preface , 2005 .

[13]  Curved exponential families of stochastic processes and their envelope families , 1996 .

[14]  Satoshi Aoki,et al.  Minimal and minimal invariant Markov bases of decomposable models for contingency tables , 2010 .

[15]  T. W. Anderson,et al.  Statistical Inference about Markov Chains , 1957 .

[16]  A. Takemura,et al.  Minimal Basis for a Connected Markov Chain over 3 × 3 ×K Contingency Tables with Fixed Two‐Dimensional Marginals , 2003 .

[17]  Seth Sullivant,et al.  A finiteness theorem for Markov bases of hierarchical models , 2007, J. Comb. Theory, Ser. A.

[18]  A. Dobra.  Markov bases for decomposable graphical models , 2003 .

[19]  M. Sørensen,et al.  Exponential Families of Stochastic Processes , 1997 .

[20]  Jeroen K. Vermunt,et al.  Modeling Joint and Marginal Distributions in the Analysis of Categorical Panel Data , 2001 .

[21]  Patsy Haccou,et al.  Statistical Analysis of Behavioural Data: An Approach Based on Time-structured Models , 1992 .