An exploratory canonical analysis approach for multinomial populations based on the $\phi$-divergence measure

Pardo, J. A.   ; Pardo, L.   ; Pardo, M. C.   ; Zografos, K.

Kybernetika, Tome 40 (2004) no. 6, pp. 757-776

Abstract (VO)

In this paper we consider an exploratory canonical analysis approach for multinomial population based on the $\phi $-divergence measure. We define the restricted minimum $\phi $-divergence estimator, which is seen to be a generalization of the restricted maximum likelihood estimator. This estimator is then used in $\phi $-divergence goodness-of-fit statistics which is the basis of two new families of statistics for solving the problem of selecting the number of significant correlations as well as the appropriateness of the model.

Détail

Classification : 62B10, 62F10, 62F30, 62H17, 62H20
Keywords: canonical analysis; restricted minimum $\phi $-divergence estimator; minimum $\phi $-divergence statistic; simulation; power divergence

Bibliographie

[1] Aitchison J., Silvey S. D.: Maximum-likelihood estimation of parameters subject to constraints. Ann. Math. Statist. 29 (1958), 813–828 | DOI | MR

[2] Ali S. M., Silvey S. D.: A general class of coefficients of divergence of one distribution from another. J. Roy. Statist. Soc. Ser. B 26 (1966), 131–142 | MR | Zbl

[3] Anderson E. B. A.: The Statistical Analysis of Categorical Data. Springer-Verlag, Berlin 1990

[4] Basu A., Basu S.: Penalized minimum disparity methods in multinomials models. Statistica Sinica 8 (1998), 841–860 | MR | Zbl

[5] Basu A., Lindsay B. G.: Minimum disparity estimation for continuous models. Ann. Inst. Statist. Math. 46 (1994), 683–705 | DOI | MR | Zbl

[6] Basu A., Sarkar S.: Minimum disparity estimation in the errors-invariables model. Statist. Probab. Lett. 20 (1994), 69–73 | DOI | MR

[7] Basu A., Sarkar S.: The trade-off between robustness and efficiency and the effect of model smoothing. J. Statist. Comput. Simul. 50 (1994), 173–185 | DOI

[8] Benzecri J. P.: L’Analyse des Données. Tome 2: L’Analyse des Correspondances. Dunod, Paris 1973 | Zbl

[9] Birch M. W.: A new proof of the Pearson–Fisher theorem. Ann. Math. Statist. 35 (1964), 817–824 | DOI | MR | Zbl

[10] Cressie N. A. C., Pardo L.: Minimum $\phi $-divergence estimator and hierarchical testing in loglinear models. Statistica Sinica 10 (2000), 867–884 | MR | Zbl

[11] Cressie N. A. C., Pardo L.: Model checking in loglinear models using $\phi $-divergences and MLEs. J. Statist. Plann. Inference 103 (2002), 437–453 | DOI | MR | Zbl

[12] Cressie N. A. C., Read T. R. C.: Multinomial goodness-of-fit tests. J. Roy. Statist. Soc. Ser. B 46 (1984), 440–464 | MR | Zbl

[13] Crichton N. J., Hinde J. P.: Correspondence analysis as a screening method for indicants for clinical diagnosis. Statistics in Medicine 8 (1989), 1351–1362 | DOI

[14] Csiszár I.: Eine Informationstheoretische Ungleichung und ihre Anwendung auf den Beweis der Ergodizität on Markhoffschen Ketten. Publ. Math. Inst. Hungar. Acad. Sci. Ser. A 8 (1963), 85–108 | MR

[15] Dahdouh B., Durantan J. F., Lecoq M.: Analyse des donnée sur l’ecologie des acridients d’Afrique de lóuest. Cahiers de l’ Analyse des Données 3 (1978), 459–482

[16] Fasham M. J. R.: A comparison of nonmetric multidimensional scaling, principal components averaging for the ordination of simulated coenocicles, and coenoplanes. Ecology 58 (1977), 551–561 | DOI

[17] Gilula Z., Haberman J.: Canonical Analysis of Contingency Tables by Maximum Likelihood. J. Amer. Statist. Assoc. 81 (1986), 395, 780–788 | DOI | MR | Zbl

[18] Greenacre M. J.: Theory and Applications of Correspondence Analysis. Academic Press, New York 1984 | MR | Zbl

[19] Greenacre M.: Correspondence analysis in medical research. Statist. Meth. Medic. Res. 1 (1992), 97–117 | DOI

[20] Greenacre M. J.: Correspondence Analysis in Practice. Academic Press, London 1993 | Zbl

[21] Greenacre M. J.: Correspondence Analysis of the Spanish National Health Survey. Department of Economics and Business, Universitat Pompeu Fabra, Barcelona 2002

[22] Lancaster H. O.: The Chi-squared Distribution. Wiley, New York 1969 | MR | Zbl

[23] Lebart L., Morineau, A., Warwick K.: Multivariate Descriptive Statistical Analysis. Wiley, New York 1984 | MR | Zbl

[24] Lindsay B. G.: Efficiency versus robustness. The case for minimum Hellinger distance and other methods. Ann. Statist. 22 (1994), 1081–1114 | DOI | MR | Zbl

[25] Matthews G. B., Crowther N. A. S.: A maximum likelihood estimation procedure when modelling categorical data in terms of cross-product ratios. South African Statist. J. 31 (1997), 161–184 | MR | Zbl

[26] Matthews G. B., Crowther N. A. S.: A maximum likelihood estimation procedures when modeling in terms of constraints. South African Statist. J. 29 (1995), 29–50 | MR

[27] Morales D., Pardo, L., Vajda I.: Asymptotic divergence of estimates of discrete distributions. J. Statist. Plann. Inference 48 (1995), 347–369 | DOI | MR | Zbl

[28] Parr W. C.: Minimum distance estimation: a bibliography. Comm. Statist. Theory Methods 10 (1981), 1205–1224 | DOI | MR | Zbl

[29] Pardo J. A., Pardo, L., Zografos K.: Minimum $\phi $-divergence estimators with constraints in multinomial populations. J. Statist. Plann. Inference 104 (2002), 221–237 | DOI | MR | Zbl

[30] Read T. R. C., Cressie N. A. C.: Goodness-of-fit Statistics for Discrete Multivariate Data. Springer, New York 1988 | MR | Zbl

[31] Srole L., Langner T. S., Michael S. T., Opler M. K., Reannie T. A. C.: Mental Health in the Metropolis: The Midtown Manhattan Study. McGraw-Hill, New York 1962

[32] Wolfowitz J.: Estimation by minimum distance method. Ann. Inst. Statist. Math. 5 (1953), 9–23 | DOI | MR