## Context-specific independence in graphical log-linear models.(English)Zbl 1348.65036

Summary: Log-linear models are the popular workhorses of analyzing contingency tables. A log-linear parameterization of an interaction model can be more expressive than a direct parameterization based on probabilities, leading to a powerful way of defining restrictions derived from marginal, conditional and context-specific independence. However, parameter estimation is often simpler under a direct parameterization, provided that the model enjoys certain decomposability properties. Here we introduce a cyclical projection algorithm for obtaining maximum likelihood estimates of log-linear parameters under an arbitrary context-specific graphical log-linear model, which needs not satisfy criteria of decomposability. We illustrate that lifting the restriction of decomposability makes the models more expressive, such that additional context-specific independencies embedded in real data can be identified. It is also shown how a context-specific graphical model can correspond to a non-hierarchical log-linear parameterization with a concise interpretation. This observation can pave way to further development of non-hierarchical log-linear models, which have been largely neglected due to their believed lack of interpretability.

### MSC:

 65C60 Computational problems in statistics (MSC2010) 62H17 Contingency tables 62A09 Graphical methods in statistics
Full Text:

### References:

 [1] Corander, J, Labelled graphical models, Scand J Stat, 30, 493-508, (2003) · Zbl 1034.62049 [2] Corander, J; Gyllenberg, M; Koski, T, Bayesian model learning based on a parallel MCMC strategy, Stat Comput, 16, 355-362, (2006) [3] Corander, J; Ekdahl, M; Koski, T, Parallel interacting MCMC for learning of topologies of graphical models, Data Min Knowl Disc, 17, 431-456, (2008) [4] Csiszár, I, $$I$$-divergence geometry of probability distributions and minimization problems, Ann Probab, 3, 146-158, (1975) · Zbl 0318.60013 [5] Csiszár, I; Matús̆, F, Information projections revisited, IEEE Trans Inf Theory, 49, 1474-1490, (2003) · Zbl 1063.94016 [6] Edwards, D; Havránek, T, A fast procedure for model search in multidimensional contingency tables, Biometrika, 72, 339-351, (1985) · Zbl 0576.62067 [7] Eriksen PS (1999) Context specific interaction models. Technical report, Department of Mathematical Sciences, Aalborg University, Aalborg · Zbl 1051.62053 [8] Friedman N, Goldszmidt M (1996) Learning Bayesian networks with local structure. In: Proceedings of the twelfth annual conference on uncertainty in artificial intelligence, pp 252-262 · Zbl 0910.68176 [9] Golumbic MC (2004) Algorithmic graph theory and perfect graphs, 2nd edn. Elsevier, Amsterdam · Zbl 1050.05002 [10] Helsingin Sanomat (2011) HS:n vaalikone 2011. http://www2.hs.fi/extrat/hsnext/HS-vaalikone2011.xls. Visited 19 Aug 2014 · Zbl 0576.62067 [11] Højsgaard, S, Split models for contingency tables, Comput Stat Data Anal, 42, 621-645, (2003) · Zbl 1429.62202 [12] Højsgaard, S, Statistical inference in context specific interaction models for contingency tables, Scand J Stat, 31, 143-158, (2004) · Zbl 1051.62053 [13] Koller D, Friedman N (2009) Probabilistic graphical models: principles and techniques. MIT Press, London · Zbl 1183.68483 [14] Lauritzen SL (1996) Graphical models. Oxford University Press, Oxford · Zbl 0907.62001 [15] Nyman, H; Pensar, J; Koski, T; Corander, J, Stratified graphical models—context-specific independence in graphical models, Bayesian Anal, 9, 883-908, (2014) · Zbl 1327.62030 [16] Nyman H, Xiong J, Pensar J, Corander J (2015) Marginal and simultaneous predictive classification using stratified graphical models. Adv Data Anal Classif. doi:10.1007/s11634-015-0199-5 · Zbl 1414.62262 [17] Pensar, J; Nyman, H; Koski, T; Corander, J, Labeled directed acyclic graphs: a generalization of context-specific independence in directed graphical models, Data Min Knowl Disc, 29, 503-533, (2015) · Zbl 1403.68206 [18] Rudas, T, A new algorithm for the maximum likelihood estimation of graphical log-linear models, Comput Stat, 13, 529-537, (1998) · Zbl 0923.62057 [19] Schwarz, G, Estimating the dimension of a model, Ann Stat, 6, 461-464, (1978) · Zbl 0379.62005 [20] Whittaker J (1990) Graphical models in applied multivariate statistics. Wiley, Chichester · Zbl 0732.62056
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.