Abstract
Graphical Markov models are multivariate statistical models in which the joint distribution satisfies independence statements that are captured by a graph. We consider models for discrete variables, that are analogous to multivariate regressions in the linear case, when the variables can be arranged in sequences of joint response, intermediate and purely explanatory variables. In the case of one single group of variables, these models specify marginal independencies of pairs of variables. We show that the models admit a proper marginal log-linear parameterization that can accommodate all the marginal and conditional independence constraints involved, and can be fitted, using maximum likelihood under a multinomial assumption, by a general iterative gradient-based algorithm. We discuss a technique for determining fast approximate estimates, that can also be used for initializing the general algorithm and we present an illustration based on data from the U.S. General Social Survey.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
AITCHISON, J. and SILVEY, S.D. (1958): Maximum likelihood estimation of parameter subject to restraints. Annals of Mathematical Statistics 29, 813–828.
BARTOLUCCI, F., COLOMBI, R. and FORCINA, A. (2007): An extended class of marginal link functions for modelling contingency tables by equality and inequality constraints. Statistica Sinica 17, 691–711.
BERGSMA, W.P. (1997): Marginal models for categorical data, Ph.D Thesis, Tilburg.
BERGSMA, W.P. and RUDAS, T. (2002): Marginal log-linear models for categorical data. Annals of Statistics 30, 140–159.
COX, D.R. (2006): Principles of Statistical Inference. Cambridge University Press, Cambridge.
COX, D.R. and WERMUTH, N. (1990): An approximation to maximum likelihood estimates in reduced models. Biometrika 77, 747–761.
COX, D.R. and WERMUTH, N. (1996): Multivariate dependencies. Models, analysis and interpretation. Chapman and Hall, London.
DAVIS, J.A., SMITH, T.W. and MARDSEN, J.A. (2007): General Social Surveys Cumulative Codebook: 1972-2006. NORC, Chicago. URL http://sda.berkeley.edu/GSS/.
DRTON, M., EICHLER, M. and RICHARDSON, T.S. (2007): Identification and Likelihood Inference for Recursive Linear Models with Correlated Errors. arXiv:0601631 .
DRTON, M. and RICHARDSON, T.S. (2008): Binary models for marginal independence. J. Royal Statist. Soc., Ser. B 70, 287–309.
GLONEK, G.F.V. and McCULLAGH, P. (1995): Multivariate logistic models. J. Royal Statist. Soc., Ser. B 57, 533–546.
KAUERMANN, G. (1996): On a dualization of graphical Gaussian models. Scand. J. of Statistics 23, 105–116.
KAUERMANN, G. (1997): A note on multivariate logistic models for contingency tables. Australian Journal of Statistics 39(3), 261–276.
LAURITZEN, S.L. (1996): Graphical Models. Oxford University Press, Oxford.
LUPPARELLI, M., MARCHETTI, G.M. and BERGSMA, W.P. (2008): Parameterizations and fitting of bi-directed graph models to categorical data. arXiv:0801.1440 .
R DEVELOPMENT CORE TEAM (2008): R: A language and environment for sta- tistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org.
RICHARDSON, T.S. (2003): Markov property for acyclic directed mixed graphs. Scand. J. of Statistics 30, 145–157.
RODDAM, A.W. (2004): An approximate maximum likelihood procedure for pa- rameter estimation in multivariate discrete data regression models. J. of Applied Statistics 28, 273–279.
RUDAS, T., BERGSMA, W.P. and NÉMETH, R. (2006): Parameterization and estimation of path models for categorical data. In: A. Rizzi & M. Vichi, (Eds.): Compstat 2006 Proceedings. Physica-Verlag, Heidelberg, 383–394.
WERMUTH, N. and COX, D.R. (2004): Joint response graphs and separation induced by triangular systems. J. Royal Statist. Soc. B 66, 687–717.
WERMUTH, N., COX, D.R. and MARCHETTI, G.M. (2006): Covariance chains. Bernoulli, 12, 841–862.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2008 Physica-Verlag Heidelberg
About this paper
Cite this paper
Marchetti, G.M., Lupparelli, M. (2008). Parameterization and Fitting of a Class of Discrete Graphical Models. In: Brito, P. (eds) COMPSTAT 2008. Physica-Verlag HD. https://doi.org/10.1007/978-3-7908-2084-3_10
Download citation
DOI: https://doi.org/10.1007/978-3-7908-2084-3_10
Publisher Name: Physica-Verlag HD
Print ISBN: 978-3-7908-2083-6
Online ISBN: 978-3-7908-2084-3
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)