Source: Ann. Statist. Volume 31, Number 3
(2003), 807-832.
This paper describes the large sample properties of the likelihood ratio test statistic (LRTS) when the parameters characterizing the true null distribution are not unique. It is well known that the classical asymptotic theory for the likelihood ratio test does not apply to such problems and the LRTS may not have the typical chi-squared type limiting distribution. This paper establishes a general quadratic approximation of the log-likelihood ratio function in a Hellinger neighborhood of the true density which is valid with or without loss of identifiability of the true distribution. Under suitable conditions, the asymptotic null distribution of the LRTS under loss of identifiability can be obtained by maximizing the quadratic form. These results extend the work of Chernoff and Le Cam. In particular, applications to testing the number of mixture components in finite mixture models are discussed.
References
AZAÏS, J.-M., CIERCO-Ay ROLLES, C. and CROQUETTE, A. (1999). Bounds and asy mptotic expansions for the distribution of the maximum of a smooth stationary Gaussian process. ESAIM Probab. Statist. 3 107-129.
AZAÏS, J.-M. and WSCHEBOR, M. (2000). Computing the distribution of the maximum of a Gaussian process. Univ. Sabatier, Toulouse. Preprint.
CHERNOFF, H. (1954). On the distribution of the likelihood ratio. Ann. Math. Statist. 25 573-578.
CHERNOFF, H. and LANDER, E. (1995). Asy mptotic distribution of the likelihood ratio test that a mixture of two binomials is a single binomial. J. Statist. Plann. Inference 43 19-40.
DACUNHA-CASTELLE, D. and GASSIAT, E. (1997). Testing in locally conic models and application to mixture models. ESAIM Probab. Statist. 1 285-317.
DACUNHA-CASTELLE, D. and GASSIAT, E. (1999). Testing the order of a model using locally conic parameterization: Population mixtures and stationary ARMA processes. Ann. Statist. 27 1178-1209.
DUDLEY, R. M. (1999). Uniform Central Limit Theorems. Cambridge Univ. Press.
LE CAM, L. (1970). On the assumptions used to prove asy mptotic normality of maximum likelihood estimates. Ann. Math. Statist. 41 802-828.
LEMDANI, M. and PONS, O. (1999). Likelihood ratio tests in contamination models. Bernoulli 5 705-719.
LINDSAY, B. G. (1995). Mixture Models: Theory, Geometry and Applications. IMS, Hay ward, CA.
MCLACHLAN, G. J. and BASFORD, K. E. (1988). Mixture Models: Inference and Applications to Clustering. Dekker, New York.
Mathematical Reviews (MathSciNet):
MR926484
PRAKASA RAO, B. L. S. (1992). Identifiability in Stochastic Models: Characterization of Probability Distributions. Academic Press, London.
REDNER, R. A. (1981). Note on the consistency of the maximum likelihood estimate for nonidentifiable distributions. Ann. Statist. 9 225-228.
TITTERINGTON, D. M., SMITH, A. F. M. and MAKOV, U. E. (1985). Statistical Analy sis of Finite Mixture Distributions. Wiley, New York.
VAN DER VAART, A. W. and WELLNER, J. A. (1996). Weak Convergence and Empirical Processes with Applications to Statistics. Springer, New York.
NEW YORK, NEW YORK 10021 E-MAIL: liuxin@linkage.rockefeller.edu DEPARTMENT OF STATISTICS COLUMBIA UNIVERSITY 2990 BROADWAY
NEW YORK, NEW YORK 10027 E-MAIL: yshao@stat.columbia.edu
