Statistical inference in constrained latent class models for multinomial data based on $$\phi $$ϕ-divergence measures

In this paper we explore the possibilities of applying $$\phi $$ϕ-divergence measures in inferential problems in the field of latent class models (LCMs) for multinomial data. We first treat the problem of estimating the model parameters. As explained below, minimum $$\phi $$ϕ-divergence estimators (M$$\phi $$ϕEs) considered in this paper are a natural extension of the maximum likelihood estimator (MLE), the usual estimator for this problem; we study the asymptotic properties of M$$\phi $$ϕEs, showing that they share the same asymptotic distribution as the MLE. To compare the efficiency of the M$$\phi $$ϕEs when the sample size is not big enough to apply the asymptotic results, we have carried out an extensive simulation study; from this study, we conclude that there are estimators in this family that are competitive with the MLE. Next, we deal with the problem of testing whether a LCM for multinomial data fits a data set; again, $$\phi $$ϕ-divergence measures can be used to generate a family of test statistics generalizing both the classical likelihood ratio test and the chi-squared test statistics. Finally, we treat the problem of choosing the best model out of a sequence of nested LCMs; as before, $$\phi $$ϕ-divergence measures can handle the problem and we derive a family of $$\phi $$ϕ-divergence test statistics based on them; we study the asymptotic behavior of these test statistics, showing that it is the same as the classical test statistics. A simulation study for small and moderate sample sizes shows that there are some test statistics in the family that can compete with the classical likelihood ratio and the chi-squared test statistics.

[1]  Clifford C. Clogg,et al.  Latent Class Models for Measuring , 1988 .

[2]  I. Vajda,et al.  Asymptotic divergence of estimates of discrete distributions , 1995 .

[3]  Leandro Pardo,et al.  Phi-Divergence test statistics for testing the validity of latent class models for binary data , 2014 .

[4]  L. A. Goodman.  Exploratory latent structure analysis using both identifiable and unidentifiable models , 1974 .

[5]  Paul P. Biemer,et al.  Latent Class Analysis of Survey Error , 2011 .

[6]  M. W. Birch.  A New Proof of the Pearson-Fisher Theorem , 1964 .

[7]  Francesco Bartolucci,et al.  Extended RC Association Models Allowing for Order Restrictions and Marginal Modeling , 2002 .

[8]  Paul F. Lazarsfeld,et al.  Latent Structure Analysis. , 1969 .

[9]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[10]  Francesco Bartolucci,et al.  A multilevel finite mixture item response model to cluster examinees and schools , 2014, Adv. Data Anal. Classif..

[11]  Leandro Pardo,et al.  Phi‐Divergence Statistic , 2006 .

[12]  A. Agresti.  An introduction to categorical data analysis , 1997 .

[13]  Anton K. Formann,et al.  Linear Logistic Latent Class Analysis , 1982 .

[14]  Leandro Pardo,et al.  Comparing two treatments in terms of the likelihood ratio order , 2014, 1402.6574.

[15]  Peter M. Bentler,et al.  Ensuring Positiveness of the Scaled Difference Chi-square Test Statistic , 2008, Psychometrika.

[16]  L. Pardo.  Statistical Inference Based on Divergence Measures , 2005 .

[17]  Keryn E. Pasch,et al.  Latent Class Analysis of Lifestyle Characteristics and Health Risk Behaviors among College Youth , 2009, Prevention Science.

[18]  Z. Q. Lu.  Statistical Inference Based on Divergence Measures , 2007 .

[19]  A Felipe,et al.  Minimum {\phi}-divergence estimation in constrained latent class models for binary data , 2014, 1406.0109.

[20]  J. Hagenaars,et al.  Applied Latent Class Analysis , 2003 .

[21]  Ewa Genge,et al.  A latent class analysis of the public attitude towards the euro adoption in Poland , 2014, Adv. Data Anal. Classif..

[22]  J. Berkson.  MINIMUM CHI-SQUARE, NOT MAXIMUM LIKELIHOOD! , 1980 .

[23]  Fred B. Bryant,et al.  Principles and Practice of Scaled Difference Chi-Square Testing , 2012 .

[24]  Anton K. Formann,et al.  Constrained latent class models: Theory and applications , 1985 .

[25]  Daniel L. Oberski.  Beyond the number of classes: separating substantive from non-substantive dependence in latent class analysis , 2015, Advances in Data Analysis and Classification.

[26]  A. Formann.  Linear Logistic Latent Class Analysis for Polytomous Data , 1992 .

[27]  Leandro Pardo,et al.  Size and power considerations for testing loglinear models using divergence test statistics , 2003 .

[28]  Beau Abar,et al.  Self-regulated learning and self-directed study in a pre-college sample. , 2010, Learning and individual differences.

[29]  Timothy R. C. Read,et al.  Multinomial goodness-of-fit tests , 1984 .

[30]  Katherine E. Masyn,et al.  New approaches to studying problem behaviors: a comparison of methods for modeling longitudinal, categorical adolescent drinking data. , 2009, Developmental psychology.

[31]  L. A. Goodman.  Simple Models for the Analysis of Association in Cross-Classifications Having Ordered Categories , 1979 .

[32]  Leandro Pardo,et al.  Minimum phi divergence estimator and hierarchical testing in loglinear models , 2000 .

[33]  J. Dale,et al.  Asymptotic Normality of Goodness‐Of‐Fit Statistics for Sparse Product Multinomials , 1986 .