Finite mixture models for proportions.

Six data sets recording fetal control mortality in mouse litters are presented. The data are clearly overdispersed, and a standard approach would be to describe the data by means of a beta-binomial model or to use quasi-likelihood methods. For five of the examples, we show that beta-binomial model provides a reasonable description but that the fit can be significantly improved by using a mixture of a beta-binomial model with a binomial distribution. This mixture provides two alternative solutions, in one of which the binomial component indicates a high probability of death but is selected infrequently; this accounts for outlying litters with high mortality. The influence of the outliers on the beta-binomial fits is also demonstrated. The location and nature of the two main maxima to the likelihood are investigated through profile log-likelihoods. Comparisons are made with the performance of finite mixtures of binomial distributions.

[1]  J K Haseman,et al.  The distribution of fetal death in control mice and its implications on statistical tests for dominant lethal effects. , 1976, Mutation research.

[2]  H. Akaike,et al.  Information Theory and an Extension of the Maximum Likelihood Principle , 1973 .

[3]  Byung Soo Kim,et al.  The Ames Salmonella/Microsome Mutagenicity Assay: Issues of Inference and Validation , 1989 .

[4]  M. Puterman,et al.  Mixed Poisson regression models with covariate dependent rates. , 1996, Biometrics.

[5]  L Vuataz,et al.  Use of the beta-binomial distribution in dominant-lethal testing for "weak mutagenic activity: part 2. , 1977, Mutation research.

[6]  Ross L. Prentice,et al.  Binary Regression Using an Extended Beta-Binomial Distribution, with Discussion of Correlation Induced by Covariate Measurement Errors , 1986 .

[7]  Byron J. T. Morgan,et al.  Automatic starting point selection for function optimization , 1994 .

[8]  J. W. Akitt.  Function Minimisation Using the Nelder and Mead Simplex Method with Limited Arithmetic Precision: The Self Regenerative Simplex , 1977, Comput. J..

[9]  M. Puterman,et al.  Maximum-penalized-likelihood estimation for independent and Markov-dependent mixture models. , 1992, Biometrics.

[10]  L L Kupper,et al.  The use of a correlated binomial model for the analysis of certain toxicological experiments. , 1978, Biometrics.

[11]  M. Gerson.  The Techniques and Uses of Probability Plotting , 1975 .

[12]  J. Neyman,et al.  Outline of statistical treatment of the problem of diagnosis. , 1947, Public health reports.

[13]  H. Tong.  Non-linear time series. A dynamical system approach , 1990 .

[14]  G. Schwarz.  Estimating the Dimension of a Model , 1978 .

[15]  G. McLachlan,et al.  Likelihood Estimation with Normal Mixture Models , 1985 .

[16]  B. J. Morgan,et al.  Modelling digit preference in fecundability studies. , 1991, Biometrics.

[17]  K. Liang,et al.  On the use of the quasi-likelihood method in teratological experiments. , 1994, Biometrics.

[18]  G. McLachlan.  On Bootstrapping the Likelihood Ratio Test Statistic for the Number of Components in a Normal Mixture , 1987 .

[19]  D. James,et al.  Analysis of results from a collaborative study of the dominant lethal assay. , 1982, Mutation research.

[20]  D. Schwartz,et al.  Sterility and fecundability estimation. , 1983, Journal of theoretical biology.

[21]  S. Raul.  On the beta-correlated binomial (bcb) distribution - a three parameter generalization of the binomial distribution , 1987 .

[22]  P. Altham,et al.  Two Generalizations of the Binomial Distribution , 1978 .

[23]  Peter J. Danaher,et al.  PARAMETER ESTIMATION AND APPLICATIONS FOR A GENERALISATION OF THE BETA‐BINOMIAL DISTRIBUTION , 1988 .

[24]  Bruce G. Lindsay,et al.  Computer-assisted analysis of mixtures (C.A.MAN) statistical algorithms , 1992 .

[25]  M. Repetto,et al.  A combined strategy for optimization in nonlinear magnetic problems using simulated annealing and search techniques , 1992 .

[26]  Murray Aitkin,et al.  Statistical Modelling of Data on Teaching Styles , 1981 .

[27]  Anthony C. Atkinson,et al.  A Method for Discriminating between Models , 1970 .

[28]  Byron J. T. Morgan.  Analysis of Quantal Response Data , 1992 .

[29]  J. Simkin,et al.  Optimizing electromagnetic devices combining direct search methods with simulated annealing , 1992 .

[30]  D. Rubin,et al.  Estimation and Hypothesis Testing in Finite Mixture Models , 1985 .