×

Weighted least-squares and quasilikelihood estimation for categorical data under singular models. (English) Zbl 0698.62071

Summary: Analysis of categorical variables by generalized linear models having singular covariance matrices is considered. A weighted least-squares estimator is proposed, and is shown to be “asymptotically best linear unbiased” under general sampling schemes. This estimator is also shown to be equivalent to estimators obtained from two other weighted least- squares approaches. Finally, a “quasilikelihood” estimator is proposed for special covariance structures, which include product multinomial sampling and Dirichlet-multinomial models for two-stage cluster sampling. This is obtained directly without having to take explicit account of the sampling restrictions on the parameters. As a corollary, the “asymptotically best linear unbiased estimator” is shown to be “best asymptotically normal” for product multinomial sampling. Large-sample tests of goodness of fit and of hypotheses on the model parameters, and examples of applications of the results, are also presented.

MSC:

62J99 Linear inference, regression
62H12 Estimation in multivariate analysis
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Apostol, T. M., Mathematical Analysis: A Modern Approach to Advanced Calculus (1957), Addison-Wesley: Addison-Wesley Reading, Mass · Zbl 0077.05501
[2] Baker, R.; Nelder, J. A., General Linear Interactive Modelling (GLIM) (1978), Numerical Algorithms Group: Numerical Algorithms Group Oxford, Release 3
[3] Birch, M. W., Maximum likelihood in three-way contingency tables, J. Roy. Statist. Soc. Ser. B, 25, 220-233 (1963) · Zbl 0121.14001
[4] Brier, S. S., Analysis of contingency tables under cluster sampling, Biometrika, 67, 591-596 (1980) · Zbl 0455.62045
[5] Fay, R. E., A jack-knifed chisquared test for complex samples, J. Amer. Statist. Assoc., 80, 148-157 (1985)
[6] Grizzle, J. E.; Starmer, C. F.; Koch, G. G., Analysis of categorical data by linear models, Biometrics, 25, 489-504 (1969) · Zbl 1149.62317
[7] Grizzle, J. E.; Williams, O. D., Log-linear models and tests of independence for contingency tables, Biometrics, 28, 137-156 (1972)
[8] Haber, M., Maximum likelihood methods for linear and log-linear models in categorical data analysis, Comput. Statist. Data Anal., 3, 1-10 (1985) · Zbl 0586.62084
[9] Hidiroglou, M. A.; Rao, J. N.K., Chi-squared tests with categorical data from complex surveys, Part I, J. Official Statist., 3, 117-132 (1987)
[10] Jorgensen, B., Maximum likelihood estimation and large-sample inference for generalized linear and non-linear regression models, Biometrika, 70, 19-28 (1983) · Zbl 0526.62067
[11] McCullagh, P., Quasi-likelihood functions, Ann. Statist., 11, 59-67 (1983) · Zbl 0507.62025
[12] McCullagh, P.; Nelder, J. A., Generalized Linear Models (1983), Chapman and Hall: Chapman and Hall London · Zbl 0588.62104
[13] Mitra, S. K., Unified least squares approach to linear estimation in a generalized Gauss-Markov model, SIAM J. Appl. Math., 35, 671-680 (1973) · Zbl 0244.62054
[14] Prentice, R. L.; Pyke, R., Logistic disease incidence models and case-control studies, Biometrika, 66, 403-411 (1979) · Zbl 0428.62078
[15] Rao, C. R., Unified theory of linear estimation, Sankhyá, 33, 371-394 (1971) · Zbl 0236.62048
[16] Rao, C. R., Linear Statistical Inference and its Applications (1973), Wiley: Wiley New York · Zbl 0169.21302
[17] Stirling, W. D., Testing linear hypotheses in contingency tables with zero cell counts, Comput. Statist. Data Anal., 4, 1-10 (1986) · Zbl 0588.62091
[18] Thomas, D. R.; Rao, J. N.K., Small-sample comparisons of level and power for simple goodness-of-fit statistics under cluster sampling, J. Amer. Statist. Assoc., 82, 630-636 (1987)
[19] Thomas, D. R.; Rao, J. N.K., On the analysis of categorical variables by linear models having singular dispersion matrices, Biometrics, 44, 243-248 (1988) · Zbl 0707.62141
[20] Woolson, R. F.; Brier, S. S., Equivalence of certain chi-squared test statistics, Amer. Statist., 35, 250-254 (1981)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.