Abstract
Marginal models involve restrictions on the conditional and marginal association structure of a set of categorical variables. They generalize log-linear models for contingency tables, which are the fundamental tools for modelling the conditional association structure. This chapter gives an overview of the development of marginal models during the past 20 years. After providing some motivating examples, the first few sections focus on the definition and characteristics of marginal models. Specifically, we show how their fundamental properties can be understood from the properties of marginal log-linear parameterizations. Algorithms for estimating marginal models are discussed, focussing on the maximum likelihood and the generalized estimating equations approaches. It is shown how marginal models can help to understand directed graphical and path models, and a description is given of marginal models with latent variables.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The design also has disadvantages, of course. These include panel attrition and the fact that, even if originally selected appropriately, with passing time the sample will become different in composition from the current population.
- 2.
See Bergsma et al. [12], Chapter 5, for applications in causal analysis and the relationship with structural equation models.
- 3.
The neighbours of a node A are those nodes with which A is connected by an edge.
- 4.
For alternative parameterizations see Sect. 3.5.
- 5.
Note that formula (iii) in Theorem 3.1 in Ghosh and Vellaisamy [38] appears to have a typo.
- 6.
A model is called smooth if it admits a smooth parameterization.
- 7.
The notations for the variables are going to be clarified later.
- 8.
Setting a parameter to zero means setting it zero for all category combinations of the variables in the effect.
References
Agresti, A.: Categorical Data Analysis, 3rd edn. Wiley, London (2013)
Andersson, S.A., Madigan, D., Perlman, M.D.: Alternative Markov properties for chain graphs. Scand. J. Stat. 28, 33–85 (2001)
Aitchison, J., Silvey, S.D.: Maximum likelihood estimation of parameters subject to restraints. Ann. Math. Stat. 29, 813–828 (1958)
Barndorff-Nielsen, O.: Information and Exponential Families. Wiley, New York (1978)
Bartolucci, F., Forcina, A.: A class of latent marginal models for capture–recapture data with continuous covariates. J. Am. Stat. Assoc. 101, 786–794 (2006)
Bartolucci, F., Colombi, R., Forcina, A.: An extended class of marginal link functions for modelling contingency tables by equality and inequality constraints. Stat. Sin. 691–711 (2007)
Bartolucci, F., Scaccia, L., Farcomeni, A.: Bayesian inference through encompassing priors and importance sampling for a class of marginal models for categorical data. Comput. Stat. Data Anal. 56, 4067–4080 (2012)
Bergsma, W.P. (1997). Marginal models for categorical data. Tilburg: Tilburg University Press
Bergsma, W., Rudas, T.: Marginal models for categorical data. Ann. Stat. 30, 140–159 (2002)
Bergsma, W., Rudas, T.: Variation independent parameterizations of multivariate categorical distributions. In: Cuadras, C.M., Fortiana, J., Rodriguez-Lallena, J.A. (eds.) Distributions with Given Marginals and Statistical Modelling, pp. 21–27. Kluwer, Dordecht (2002)
Bergsma, W., Rudas, T.: On conditional and marginal association. Annales de la Faculte des Sciences de Toulouse 6(11), 455–468 (2003)
Bergsma, W., Croon, M., Hagenaars, J.A.: Marginal Models For Dependent, Clustered and Longitudinal Categorical Data. Springer, New York (2009)
Bergsma, W.P., Croon, M.A., Hagenaars, J.A. (2013). Advancements in marginal modeling for categorical data. Sociol. Methodol. 43(1), 141
Bergsma, W.P., Rapcsák, T.: An exact penalty method for smooth equality constrained optimization with application to maximum likelihood estimation. Eurandom Technical Report (2005)
Bishop, Y.M.M., Fienberg, S.E., Holland, P.W.: Discrete Multivariate Analysis: Theory and Practice. MIT Press, Cambridge (1975)
Bon, J., Baffour, B., Spallek, M., Haynes, M.: Analysing sensitive data from dynamically-generated overlapping contingency tables. J. Off. Stat. 36, 275–296 (2020)
Cocchi, M. (ed.): Data Fusion Methodology and Applications. Elsevier, Amsterdam (2019)
Colombi, R., Forcina, A.: Marginal regression models for the analysis of positive association of ordinal response variables. Biometrika 88, 1007–1019 (2001)
Colombi, R., Forcina, A.: A class of smooth models satisfying marginal and context specific conditional independencies. J. Multivariate Anal. 126, 75–85 (2014)
Colombi, R., Giordano, S.: Multiple hidden Markov models for categorical time series. J. Multivariate Anal. 140, 19–30 (2015)
Colombi, R., Forcina, A.: Testing order restrictions in contingency tables. Metrika 79, 73–90 (2016)
Colombi, R., Giordano, S., Cazzaro, M.: hmmm: an R package for hierarchical multinomial marginal models. J. Stat. Softw. 59(11), 1–25 (2014)
Colombi, R., Giordano, S., Gottard, A., Iannario, M.: Hierarchical marginal models with latent uncertainty. Scand. J. Stat. 46, 595–620 (2019)
Cox, D.R., Wermuth, N.: Multivariate Dependencies. Chapman and Hall, London (1996)
Dardanoni, V., Fiorini, M., Forcina, A.: Stochastic monotonicity in intergenerational mobility tables. J. Appl. Econom. 27, 85–107 (2012)
D’Orazio, M., Di Zio, M., Scanu, M.: Statistical matching for categorical data: displaying uncertainty and using logical constraints. J. Off. Stat. 22, 137–157 (2006)
Drton, M.: Discrete chain graph models. Bernoulli 15(3), 736–753 (2009)
Eppmann, H., Krügener, S., Schäfer, J.: First German register based census in 2011. Allgemeines Statistisches Archiv 90(3), 465–482 (2006)
Evans, R.J.: Smoothness of marginal log-linear parameterization. Electron. J. Stat. 9, 475–491 (2015)
Evans, R.J., Forcina, A.: Two algorithms for fitting constrained marginal models. Comput. Stat. Data Anal. 66, 1–7 (2013)
Evans, R.J., Richardson, T.S.: Marginal log-linear parameters for graphical Markov models. J. R. Stat. Soc. B: Stat. Methodol. 75(4), 743 (2013)
Forcina, A.: Identifiability of extended latent class models with individual covariates. Comput. Stat. Anal. 52, 5263–5268 (2008)
Forcina, A.: Smoothness of conditional independence. J. Multivariate Anal. 106, 49–56 (2012)
Forcina, A., Kateri, M.: A new general class of RC association models: estimation and main properties. J. Multivariate Anal. 184, 1–16 (2021)
Forcina, A., Lupparelli, M., Marchetti, G.M.: Marginal parameterizations of discrete models defined by a set of conditional independencies. J. Multivariate Anal. 101, 2519–2527 (2010)
Frees, E.W, Kim, J.-S.: Panel studies. In Rudas, T. (ed.) Handbook of Probability: Theory and Applications, pp. 205–224. Sage, Thousand Oaks (2008)
Frydenberg, M.: The chain graph Markov property. Scand. J. Stat. 17, 333–353 (1990)
Ghosh, S., Vellaisamy, P.: Marginal log-linear parameters and their collapsibility for categorical data (2019). arXiv 1711.00680v4
Glonek, G.F.V.: A class of regression models for multivariate categorical responses. Biometrika 83, 15–28 (1996)
Glonek, G.F.V., McCullagh, P.: Multivariate logistic models. J. R. Stat. Soc. B 57, 533–546 (1995)
Goodman, L.A.: The analysis of multidimensional contingency tales when some variables are posterior to others: a modified path analysis approach. Biometrika 60, 179–192 (1973)
Hagenaars, J.A., Bergsma, W., Croon, M.: Nonloglinear marginal latent class models. In: Advances in Latent Class Analysis: A Festschrift in Honor of C. Mitchell Dayton, vol. 61 (2019)
Huber, P.J.: The behavior of maximum likelihood estimates under nonstandard conditions. Proc. Fifth Berkeley Symp. Math. Statist. Probab. 1 221–233 (1967)
Kuijpers, R.E., Ark, L.A., Croon, M.A.: Testing hypotheses involving Cronbach’s alpha using marginal models. Br. J. Math. Stat. Psychol. 66, 503–520 (2013)
Kuijpers, R.E., Ark, L.A., Croon, M.A.: Standard errors and confidence intervals for scalability coefficients in Mokken scale analysis using marginal models. Sociol. Methodol. 43, 42–69 (2013)
Lang, J.B.: Maximum likelihood methods for a generalized class of log-linear models. Ann. Stat. 24, 726–752 (1996)
Lang, J.B.: On the comparison of multinomial and Poisson log-linear models. J. R. Stat. Soc. B (Methodological) 58(1), 253–266 (1996b)
Lang, J.B.: Multinomial-Poisson homogeneous models for contingency tables. Ann. Stat. 32,340–383 (2004)
Lang, J.B., Agresti, A.: Simultaneously modelling the joint and marginal distributions of multivariate categorical responses. J. Am. Stat. Assoc. 89, 625–632 (1994)
Lauritzen, S.L.: Graphical Models. Clarendon Press, Oxford (1996)
Lauritzen, S.L., Dawid, A.P., Larsen, B.N., Leimer, H.-G.: Independence properties of directed Markov fields. Networks 20(5), 491–505 (1990)
Lauritzen, S.L., Wermuth, N.L Graphical models for associations between variables, some of which are qualitative and some quantitative. Ann. Stat. 17, 31–57 (1989)
Liang, K.Y., Zeger, S.L.: Longitudinal data analysis using generalized linear models. Biometrika 73, 13–22 (1996)
Lipsitz, S.R., Laird, N.M., Harrington, D.P.: Generalized estimating equations for correlated binary data: using the odds ratio as a measure of association. Biometrika 78(1), 153–160 (1991)
Little, R., Rubin, D.: Statistical Analysis with Missing Data, 3rd. edn. Wiley, New York (2019)
Lupparelli, M., Marchetti, G.M., Bergsma, W.P.: Parameterizations and fitting of bi-directed graph models to categorical data. Scand. J. Stat. 36(3), 559–576 (2009)
Lupparelli, M., Roverato, A.: Log-mean linear regression models for binary responses with an application to multimorbidity. J. R. Stat. Soc. C (Applied Statistics) 66(2), 227–252 (2017)
Marchetti, G.M., Lupparelli, M.: Chain graph models of multivariate regression type for categorical data. Bernoulli 17(3), 827–844 (2011)
McCullagh, P., Nelder, J.A.: Generalized Linear Models, 2nd edn. Chapman and Hall, London (1989)
Molenberghs, G., Verbeke, G.: Models for Discrete Longitudinal Data (2005)
Németh, R., Rudas, T.: On the application of discrete marginal graphical models. Soc. Methodol. 43, 70–100 (2013)
Németh, R., Rudas, T.: Discrete graphical models in social mobility research—a comparative analysis of American, Czechoslovakian and Hungarian mobility before the collapse of state socialism. Bull. Soc. Methodol. 118, 5–21 (2013)
Nicolussi, F., Cazzaro, M.: Context-specific independencies in hierarchical multinomial marginal models. Stat. Methods Appl. 29, 767–786 (2020)
Nicolussi, F., Colombi, R.: Type II chain graph models for categorical data: a smooth subclass. Bernoulli 23, 863–883 (2017)
Ntzoufras, I., Tarantola, C., Lupparelli, M.: Probability based independence sampler for Bayesian quantitative learning in graphical log-linear marginal models. Bayesian Anal. 14, 777–803 (2019)
Pan, W.: Akaike’s information criterion in generalized estimating equations. Biometrics 57(1), 120–125 (2001)
Qaqish, B.F., Ivanova, T.: Multivariate logistic models. Biometrika 93, 1011–1017 (2006)
Rhemtulla, M., Little, T.: Tool of the trade: planned missing data designs for research in cognitive development. J. Cogn. Dev. 13(4), 10 (2012). https://doi.org/1080/15248372.2012.717340
Rotnitzky, A., Jewell, N.P.: Hypothesis testing of regression parameters in semiparametric generalized linear models for cluster correlated data. Biometrika 77(3), 485–497 (1990)
Roverato, A., Lupparelli, M., La Rocca, L.: Log-mean linear models for binary data. Biometrika 100(2), 485–494 (2013)
Rudas, T.: Prescribed conditional interaction structure models with application to the analysis of mobility tables. Q. Quantity 25, 345–358 (1991)
Rudas, T.: Odds Ratios in the Analysis of Contingency Tables. No 119, Quantitative Applications in the Social Sciences. Sage, Thousand Oaks (1998)
Rudas, T.: Informative allocation and consistent treatment selection. Stat. Methodol. 7, 323–337 (2010)
Rudas, T.: Directionally collapsible parameterizations of multivariate binary distributions. Stat. Methodol. 27, 132–145 (2015)
Rudas, T.: Lectures on Categorical Data Analysis. Springer, New York (2018)
Rudas, T., Bergsma, W.: On applications of marginal models to categorical data. Metron 42, 15–37 (2004)
Rudas, T., Bergsma, W., Németh, R.: Parameterization and estimation of path models for categorical data. In: Rizzi, A., Vichi, M. (eds.) COMPSTAT 2006, pp. 383–394. Physica Verlag, Heidelberg (2006)
Rudas, T., Bergsma, W., Németh, R.: Marginal log-linear parameterization of conditional independence models. Biometrika 97, 1006–1012 (2010)
Rudas, T., Leimer, H.-G.: Analysis of contingency tables with known conditional odds ratios or known log-linear parameters. In: Francis, B., Seeberg, G.U.H., van der Heijden, P.G.M., Jansen, W. (eds.) Statistical Modelling, pp. 313–322. Elsevier, Amsterdam (1992)
Shpitser, I., Evans, R.J., Richardson, T.S., Robins, J.M.: Sparse nested Markov models with loglinear parameters. In: Twenty-ninth Conference on Uncertainty in Artificial Intelligence, pp. 576–585 (2013)
Stanghellini, E., van der Heijden, P.G.: A multiple-record systems estimation method that takes observed and unobserved heterogeneity into account. Biometrics 60(2), 510–516 (2004)
Touloumis, A., Agresti, A., Kateri, M.: GEE for multinomial responses using a local odds ratios parameterization. Biometrics 69(3), 633–640 (2013)
Turner, E.L.: Marginal Modelling of Capture-Recapture Data. Ph.D. Thesis. McGill University Montreal (2007)
Wedderburn, R.W.: Quasi-likelihood functions, generalized linear models, and the Gauss—Newton method. Biometrika, 61(3), 439–447 (1974)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 Springer Nature Switzerland AG
About this chapter
Cite this chapter
Rudas, T., Bergsma, W. (2023). Marginal Models: An Overview. In: Kateri, M., Moustaki, I. (eds) Trends and Challenges in Categorical Data Analysis. Statistics for Social and Behavioral Sciences. Springer, Cham. https://doi.org/10.1007/978-3-031-31186-4_3
Download citation
DOI: https://doi.org/10.1007/978-3-031-31186-4_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-31185-7
Online ISBN: 978-3-031-31186-4
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)