zbMATH — the first resource for mathematics

Preliminary test estimators and phi-divergence measures in generalized linear models with binary data. (English) Zbl 1154.62051
Summary: We consider the problem of estimation of the parameters in generalized linear models (GLM) with binary data when it is suspected that the parameter vector obeys some exact linear restrictions which are linearly independent with some degree of uncertainty. Based on minimum \(\varphi\)-divergence estimation (\(M\varphi E\)), we consider some estimators for the parameters of the GLM: Unrestricted \(M\varphi E\), restricted \(M\varphi E\), Preliminary \(M\varphi E\), Shrinkage \(M\varphi E\), Shrinkage preliminary \(M\varphi E\), James-Stein \(M\varphi E\), Positive-part Stein-Rule \(M\varphi E\) and Modified preliminary \(M\varphi E\). The asymptotic bias as well as the risk with a quadratic loss function are studied under contiguous alternative hypotheses. Some discussion about dominance among the estimators studied is presented. Finally, a simulation study is carried out.

62J12 Generalized linear models (logistic models)
62F12 Asymptotic properties of parametric estimators
62F30 Parametric inference under constraints
62J07 Ridge regression; shrinkage estimators (Lasso)
62B10 Statistical aspects of information-theoretic topics
62H12 Estimation in multivariate analysis
Full Text: DOI
[1] Ali, S.M.; Silvey, S.D., A general class of coefficients of divergence of one distribution from another, J. roy. statist. soc. B, 26, 131-142, (1966) · Zbl 0203.19902
[2] Bancroft, T.A., On biases in estimation due to use of preliminary tests of significance, Ann. math. statist., 15, 190-204, (1944) · Zbl 0063.00180
[3] Cressie, N.; Read, T.R.C., Multinomial goodness-of-fit tests, J. roy. statist. soc. B, 46, 440-464, (1984) · Zbl 0571.62017
[4] I. Csiszár, Eine Informationtheorestiche Ungleichung und ihre Anwendung anf den Beweis der Ergodizität Markoffshen Ketten. Publications of the Mathematical Institute of Hungarian Academy of Sciences, Series A, 8 (1963) 84-108
[5] Ferguson, T.S., A course in large sample theory, () · Zbl 0153.47602
[6] James, W.; Stein, C., Estimation with quadratic loss, (), 361-379 · Zbl 1281.62026
[7] Kullback, S., Kullback information, (), 421-425
[8] Matin, M.A.; Md, A.K.; Saleh, E., Some improved estimators in logistic regression model, J. stat. res., 39, 2, 37-58, (2005)
[9] Matin, M.A; Md, A.K.; Saleh, E., Small-sample properties of some improved estimators in logistic regression model with skew-normally distributed explanatory variables, J. stat. res., 40, 1, 1-21, (2006)
[10] McCullagh, P.; Nelder, J.A., Generalized linear models, (1989), Chapman and Hall London · Zbl 0744.62098
[11] Pardo, L., Statistical inference based on divergence measures, (2006), Chapman & Hall/CRC New York · Zbl 1118.62008
[12] Pardo, J.A.; Pardo, M.C., Minimum phi-divergence estimators and phi-divergence statistics in GLM with binary data, Methodol. comput. appl., (2008) · Zbl 1182.62009
[13] Md, A.K.; Saleh, E.; Sen, P.K., Nonparametric estimation of location parameter after a preliminary test on regression, Ann. statist., 6, 154-168, (1978) · Zbl 0386.62036
[14] Md, A.K.; Saleh, E.; Sen, P.K., On shrinkage least-squares estimation in a parallelism problem, Comm. statist. theory methods, 15, 1451-1466, (1986) · Zbl 0611.62078
[15] Md, A.K.; Saleh, E., Theory of preliminary test and Stein-type estimation with applications, (2006), Wiley · Zbl 1094.62024
[16] Sen, P.K.; Md, A.K.; Saleh, E., On preliminary test and shrinkage M-estimation in linear models, Ann. statist., 15, 1580-1592, (1987) · Zbl 0639.62046
[17] Stein, C., Inadmissibility of the usual estimator for the Mean of a multivariate normal distribution, (), 197-206
[18] Vajda, I., Theory of statistical inference and information, (1989), Kluwer Boston · Zbl 0678.62035
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. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.