Marginal models for categorical data

By Wicher P. Bergsma, Tamas Rudas. 2002

204 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/cce5eb9cd5aa8eadd2b0b89ff3e6583ac3489b57

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

Positive Quadrant Dependence and Marginal Modeling in Two-Way Tables With Ordered Margins (2001, background): “Molenberghs and Lesaffre (1994), Glonek and McCullagh (1995), and Bergsma and Rudas (2001). By parameterizing a bivariate distribution with the global logits on each marginal distribution and the global log-odds ratios, a variety of models for the marginal distributions may be combined with restrict”
Longitudinal Data Analysis. Handbooks of Modern Statistical Methods (2009, background): “…Becker and Balagtas, 1993; Fitzmaurice, Laird, and Rotnitzky, 1993; Molenberghs and Lesaffre, 1994; Lang and Agresti, 1994; Glonek and McCullagh, 1995; Bergsma and Rudas, 2002) — none of these likelihood-based models have proven to be of real practical use except in relatively limited settings.”
Sequences of regressions and their independences (2011, background): “Discrete type IV models form also a subclass of marginal models; see Rudas, Bergsma and Nemeth (2010), Bergsma and Rudas (2002). Defining local independence statements that involve only variables in the past are equivalent to more complex local independences used by Drton (2009); see Marchetti and L”
Chain graph models of multivariate regression type for categorical data (2009, background): “The maximization of this likelihood under the above linear constraints has been discussed by several authors; see [2,3,13, 15], among others.”
An extended class of marginal link functions for modelling contingency tables by equality and inequality constraints (2007, methodology): “As in Bergsma and Rudas (2002) (hereafter BR for brevity), an interaction will be characterized by the set I of variables involved and the marginal distribution M where it is defined.”
The Use of Restricted Latent Class Models for Defining and Testing Nonparametric and Parametric Item Response Theory Models (2001, methodology): “The approach is based on work in generalized loglinear modeling (Bergsma, 1997; Lang & Agresti, 1994) and order-restricted inference with categorical variables (e.g., Dardanoni & Forcina, 1998; Robertson, Wright, & Dykstra, 1988; Vermunt, 1999).”
Binary models for marginal independence (2007, methodology): “The multivariate logistic parameterization is a special case of the marginal log-linear framework of Bergsma and Rudas (2002a). In certain cases, such as for Fig. 2(a), there may be a marginal log-linear parameterization for a marginal independence model in which the parameters are variation indepen”
The analysis of ordered categorical data: An overview and a survey of recent developments (2005, methodology): “Simulations of overdispersed count data in relatively simple structures have been conducted by Campbell et al. (1999) and Young et al.”
PROBABILITY DISTRIBUTIONS WITH SUMMARY GRAPH STRUCTURE (2010, methodology): “statistical modeling since they contain non-smooth models, such as Example 7 of Bergsma and Rudas (2002) .”
Statistical Matching for Categorical Data: Displaying Uncertainty and Using Logical Constraints (2006, background): “In fact, if the constraints are not compatible with each other, Q results in an empty set (see Bergsma and Rudas 2002, for more details and references therein).”
Discrete chain graph models (2009, background): “A probability distribution on [2](4) satisfies the two conditional independences in (6.”
Modeling Clustered Ordered Categorical Data: A Survey (2001, background): “Continuing computational advances are making ML feasible for larger problems, both for constrained ML (Bergsma, 1997; Lang, McDonald & Smith, 1999) and for maximization with respect to joint probabilities expressed in terms of the marginal model parameters and an association model (Heumann, 1997).”