Advertisement

Advances in Data Analysis and Classification

, Volume 12, Issue 3, pp 605–636 | Cite as

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

  • A. Felipe
  • N. Martín
  • P. Miranda
  • L. Pardo
Regular Article
  • 148 Downloads

Abstract

In this paper we explore the possibilities of applying \(\phi \)-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 \(\phi \)-divergence estimators (M\(\phi \)Es) 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\(\phi \)Es, showing that they share the same asymptotic distribution as the MLE. To compare the efficiency of the M\(\phi \)Es 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, \(\phi \)-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, \(\phi \)-divergence measures can handle the problem and we derive a family of \(\phi \)-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.

Keywords

Latent class models Minimum \(\phi \)-divergence estimator Maximum likelihood estimator \(\phi \)-Divergence test statistics Goodness-of-fit Nested latent class models 

Mathematics Subject Classification

Primary 62F03 Secondary 62F05 62F12 

Notes

Acknowledgements

We are very grateful to the associate editor as well as the anonymous referees for fruitful comments and remarks that have improved the final version of the paper.

References

  1. Abar B, Loken E (2010) Self-regulated learning and self-directed study in a pre-college sample. Learn Individ Differ 20:25–29CrossRefGoogle Scholar
  2. Agresti A (1996) An introduction to categorical data analysis. Wiley, New YorkzbMATHGoogle Scholar
  3. Bartolucci F, Forcina A (2002) Extended RC association models allowing for order restrictions and marginal modelling. J Am Math Assoc 97(460):1192–1199zbMATHGoogle Scholar
  4. Berkson J (1980) Minimum chi-square, not maximum likelihood! Ann Stat 8(3):457–487MathSciNetCrossRefGoogle Scholar
  5. Biemer P (2011) Latent class analysis and survey error. Wiley, New YorkzbMATHGoogle Scholar
  6. Birch M (1964) A new proof of the Pearson-Fisher theorem. Ann Math Stat 35:817–824MathSciNetCrossRefGoogle Scholar
  7. Bryant F, Satorra A (2012) Principles and practice of scaled difference chi-square testing. Struct Equ Model 19:372–398MathSciNetCrossRefGoogle Scholar
  8. Clogg C (1988) Latent class models for measuring. In: Latent trait and class models. Plenum, New York, pp 173–205CrossRefGoogle Scholar
  9. Collins L, Lanza S (2010) Latent class and latent transition analysis for the social, behavioral, and health sciences. Wiley, New YorkGoogle Scholar
  10. Cressie N, Pardo L (2000) Minimum phi-divergence estimator and hierarchical testing in loglinear models. Stat Sin 10:867–884zbMATHGoogle Scholar
  11. Cressie N, Pardo L (2002) Phi-divergence statistics. In: Elshaarawi A, Plegorich W (eds) Encyclopedia of environmetrics, vol 13. Wiley, New York, pp 1551–1555Google Scholar
  12. Cressie N, Read T (1984) Multinomial goodness-of-fit tests. J R Stat Soc Ser B 8:440–464MathSciNetzbMATHGoogle Scholar
  13. Cressie N, Pardo L, Pardo M (2003) Size and power considerations for testing loglinear models using \(\phi \)-divergence test statistics. Stat Sin 13(2):555–570MathSciNetzbMATHGoogle Scholar
  14. Csiszár I (1967) Information-type measures of difference of probability distributions and indirect observations. Stud Sci Math Hung 2:299–318MathSciNetzbMATHGoogle Scholar
  15. Dale J (1986) Asymptotic normality of goodness-of-fit statistics for sparse product multinomials. J R Stat Soc Ser B 41:48–59MathSciNetzbMATHGoogle Scholar
  16. Dempster A, Laird N, Rubin D (1977) Maximum likelihood for incomplete data via the EM algorithm. J R Stat Soc Ser B 39:1–38MathSciNetzbMATHGoogle Scholar
  17. Feldman B, Masyn K, Conger R (2009) New approaches to studying behaviors: a comparison of methods for modelling longitudinal, categorical and adolescent drinking data. Dev Psycol 45(3):652–676CrossRefGoogle Scholar
  18. Felipe A, Miranda P, Martín N, Pardo L (2014) Phi-divergence test statistics for testing the validity of latent class models for binary data. arXiv:1407.2165
  19. Felipe A, Miranda P, Pardo L (2015) Minimum \(\phi \)-divergence estimation in constrained latent class models for binary data. Psychometrika 80(4):1020–1042MathSciNetCrossRefGoogle Scholar
  20. Formann A (1982) Linear logistic latent class analysis. Biom J 24:171–190CrossRefGoogle Scholar
  21. Formann A (1985) Constrained latent class models: theory and applications. Br J Math Stat Psycol 38:87–111MathSciNetCrossRefGoogle Scholar
  22. Formann A (1992) Linear logistic latent class analysis for polytomous data. J Am Stat Assoc 87:476–486CrossRefGoogle Scholar
  23. Genge E (2014) A latent class analysis of the public attitude towards the euro adoption in Poland. Adv Data Anal Classif 8(4):427–442MathSciNetCrossRefGoogle Scholar
  24. Gnaldi M, Bacci S, Bartolucci F (2016) A multilevel finite mixture item response model to cluster examinees and schools. Adv Data Anal Classif 10(1):53–70MathSciNetCrossRefGoogle Scholar
  25. Goodman L (1974) Exploratory latent structure analysis using Goth identifiable and unidentifiable models. Biometrika 61:215–231MathSciNetCrossRefGoogle Scholar
  26. Goodman L (1979) Simple models for the analysis of association in cross-classifications having ordered categories. J Am Stat Assoc 74:537–552MathSciNetCrossRefGoogle Scholar
  27. Hagenaars JA, Cutcheon A (2002) Applied latent class analysis. Cambridge University Press, CambridgeCrossRefGoogle Scholar
  28. Laska M, Pash K, Lust K, Story M, Ehlinger E (2009) Latent class analysis of lifestyle characteristics and health risk behaviors among college youth. Prev Sci 10:376–386CrossRefGoogle Scholar
  29. Lazarsfeld P (1950) The logical and mathematical foundation of latent structure analysis. Studies in social psycology in World War II, vol IV. Princeton University Press, Princeton, pp 362–412Google Scholar
  30. Martin N, Mata R, Pardo L (2015) Comparing two treatments in terms of the likelihood ratio order. J Stat Comput Simul 85(17):3512–3534MathSciNetCrossRefGoogle Scholar
  31. Moon K, Hero A (2014) Multivariate \(f\)-divergence estimation with confidence. In: Advances in neural information processing systems, pp 2420–2428Google Scholar
  32. Morales D, Pardo L, Vajda I (1995) Asymptotic divergence of estimators of discrete distributions. J Stat Plan Inference 48:347–369CrossRefGoogle Scholar
  33. Oberski DL (2016) Beyond the number of classes: separating substantive from non-substantive dependence in latent class analysis. Adv Data Anal Classif 10(2):171–182MathSciNetCrossRefGoogle Scholar
  34. Pardo L (2006) Statistical inference based on divergence measures. Chapman and Hall CRC, Boca RatonzbMATHGoogle Scholar
  35. Satorra A, Bentler P (2010) Ensuring positiveness of the scaled difference chi-square test statistic. Psychometrika 75(2):243–248MathSciNetCrossRefGoogle Scholar
  36. Uebersax J (2009) Latent structure analysis. Web document: http://www.john-uebersax.com/stat/index.htm

Copyright information

© Springer-Verlag GmbH Germany 2017

Authors and Affiliations

  1. 1.Department of Statistics and O.R.Complutense University of MadridMadridSpain
  2. 2.Department of Statistics and O.R. II: Decision MethodsComplutense University of MadridMadridSpain

Personalised recommendations