Exactly and almost compatible joint distributions for high-dimensional discrete conditional distributions

A conditional model is a set of conditional distributions, which may be compatible or incompatible, depending on whether or not there exists a joint distribution whose conditionals match the given conditionals. In this paper, we propose a new mathematical tool called a structural ratio matrix (SRM) to develop a unified compatibility approach for discrete conditional models. With this approach, we can find all joint pdfs after confirming that the given model is compatible. In practice, it is most likely that the conditional models we encounter are incompatible. Therefore, it is important to investigate approximated joint distributions for them. We use the concept of SRM again to construct an almost compatible joint distribution, with consistency property, to represent the given incompatible conditional model.

[1]  Patrizia Berti,et al.  Compatibility results for conditional distributions , 2014, J. Multivar. Anal..

[2]  Enrique Castillo,et al.  Exact and near compatibility of discrete conditional distributions , 2002 .

[3]  Kai Wang Ng,et al.  Inversion of Bayes' Formula for Events , 2011, International Encyclopedia of Statistical Science.

[4]  Edward H Ip,et al.  Behaviour of the Gibbs sampler when conditional distributions are potentially incompatible , 2015, Journal of statistical computation and simulation.

[5]  Guo-Liang Tian,et al.  Exact statistical solutions using the inverse Bayes formulae , 2003 .

[6]  Yuchung J. Wang,et al.  Gibbs ensembles for nearly compatible and incompatible conditional models , 2011, Comput. Stat. Data Anal..

[7]  Thomas J. Jiang,et al.  COMPATIBILITY OF FINITE DISCRETE CONDITIONAL DISTRIBUTIONS , 2010 .

[8]  H. Y. Chen.  Compatibility of conditionally specified models. , 2010, Statistics & probability letters.

[9]  B. Arnold,et al.  Compatibility of partial or complete conditional probability specifications , 2004 .

[10]  B. Arnold,et al.  Conditionally Specified Distributions: An Introduction (with comments and a rejoinder by the authors) , 2001 .

[11]  B. Arnold,et al.  Compatible Conditional Distributions , 1989 .

[12]  Seth Sullivant,et al.  The space of compatible full conditionals is a unimodular toric variety , 2006, J. Symb. Comput..

[13]  Yuchung J. Wang,et al.  Comparisons of Three Approaches for Discrete Conditional Models , 2012, Commun. Stat. Simul. Comput..

[14]  Yuchung J. Wang,et al.  Canonical representation of conditionally specified multivariate discrete distributions , 2009, J. Multivar. Anal..

[15]  N. Balakrishnan,et al.  Study of incompatibility or near compatibility of bivariate discrete conditional probability distributions through divergence measures , 2015 .

[16]  Shih-Chieh Chen,et al.  On compatibility of discrete full conditional distributions: A graphical representation approach , 2014, J. Multivar. Anal..

[17]  Kun-Lin Kuo,et al.  A simple algorithm for checking compatibility among discrete conditional distributions , 2011, Comput. Stat. Data Anal..

[18]  D. V. Gokhale,et al.  Distributions most nearly compatible with given families of conditional distributions , 1998 .

[19]  S. Nadarajah,et al.  An alternative approach for compatibility of two discrete conditional distributions , 2016, 1711.00608.

[20]  M. Tan,et al.  A Unified Method for Checking Compatibility and Uniqueness for Finite Discrete Conditional Distributions , 2008 .

[21]  Kun-Lin Kuo,et al.  Compatibility of discrete conditional distributions with structural zeros , 2010, J. Multivar. Anal..