Abstract
In this paper, we explore some weakly consistent properties of quasi-maximum likelihood estimates (QMLE) concerning the quasi-likelihood equation \( \sum\nolimits_{i = 1}^n {X_i (y_i - \mu (X_i^\prime \beta ))} \) for univariate generalized linear model E(y|X) = μ(X′β). Given uncorrelated residuals {e i = Y i − μ(X ′ i β0), 1 ⩽ i ⩽ n} and other conditions, we prove that \( \hat \beta _n - \beta _0 = O_p (\underset{\raise0.3em\hbox{$\smash{\scriptscriptstyle-}$}}{\lambda } _n^{ - 1/2} ) \) holds, where \( \hat \beta _n \) is a root of the above equation, β 0 is the true value of parameter β and \( \underset{\raise0.3em\hbox{$\smash{\scriptscriptstyle-}$}}{\lambda } _n \) denotes the smallest eigenvalue of the matrix S n = ∑ n i=1 X i X ′ i . We also show that the convergence rate above is sharp, provided independent non-asymptotically degenerate residual sequence and other conditions. Moreover, paralleling to the elegant result of Drygas (1976) for classical linear regression models, we point out that the necessary condition guaranteeing the weak consistency of QMLE is S −1 n → 0, as the sample size n → ∞.
Similar content being viewed by others
References
Nelder J A, Wedderburn R W M. Generalized linear model. J Roy Statist Soc Ser A, 135(3): 370–384 (1972)
Liang K Y, Zeger S L. Longitudinal data analysis using generalized linear models. Biometrika, 73(1): 13–22 (1986)
Zeger S L, Liang K Y. Longitudinal data analysis for discrete and continuous outcomes. Biometrics, 42(1): 121–130 (1986)
Diggle P J, Liang K Y, Zeger S L. Analysis of Longitudinal Data. London: Chapman & Hall, 1994
Künsch H R. State space and hidden markov models. In: Barndorff-Nielsen O, Cox D R, Klüppelberg C, eds. Complex Stochastic Systems. London: Chapman & Hall, 2000
Andersen P K, Borgan φ, Gill R D, et al. Statistical Models Based on Counting Processes. New York: Springer-Verlag, 1993
Fahrmeir L, Kaufmann H. Consistency and asymptotic normality of the maximum likelihood estimator in generalized linear models. Ann Statist, 13(1): 342–368 (1985)
Wedderburn R W M. Quasi-likelihood functions, generalized linear models, and the Gauss-Newton method. Biometrika, 61(3): 439–447 (1974)
Collett D. Modeling Binary Data. London: Chapman & Hall, 1991
Hinde J, Démetrio C. Overdispersion: models and estimation. Comput Statist Data Anal, 27: 151–170 (1998)
Poortema K L. On modeling overdispersion of counts. Statist Neerlandica, 53(1): 5–20 (1999)
Liang K Y, McCullagh P. Case studies in binary dispersion. Biometrics, 49(2): 623–630 (1993)
Fahrmeir L. Maximum likelihood estimation in misspecified generalized linear models. Statistics, 21(4): 487–502 (1990)
Chen K, Hu I, Ying Z. Strong consistency of maximum quasi-likelihood estimators in generalized linear models with fixed and adaptive designs. Ann Statist, 27(4): 1155–1163 (1999)
Chang Y I. Strong consistency of maximum quasi-likelihood estimate in generalized linear models via a last time. Statist Probab Lett, 45: 237–246 (1999)
Chiou J M, Müller H G. Non-parametric quasi-likelihood. Ann Statist, 27(1): 36–64 (1999)
Chen X, Chen X R. Adaptive quasi-likelihood estimate in generalized linear models. Sci China Ser A-Math, 48(6): 829–846 (2005)
Yue L, Chen X R. Rate of strong consistency of quasi maximum likelihood estimate in generalized linear models. Sci China Ser A-Math, 47(6): 882–893 (2004)
Yin C M, Zhao L C. Strong consistency of maximum quasi-likelihood estimates in generalized linear models. Sci China Ser A-Math, 48(8): 1009–1014 (2005)
Gao Q B, Wu Y H. Strong consistency of quasi-maximum likelihood estimator in generalized linear regression (in Chinese). Chinese Ann Math Ser A, 25(6): 705–710 (2004)
Drygas H. Weak and strong consistency of the least squares estimators in regression models. Prob Theory Related Fields, 34(2): 119–127 (1976)
Lai T L, Robbins H, Wei C Z. Strong consistency of least squares estimates in multiple regression II. J Multivariate Anal, 9: 343–361 (1979)
Dugundji J. Topology. Boston: Allyn & Bacon, 1966
Fitzpatrick P M. Advanced Calculus: A Course in Mathematical Analysis. Boston: PWS Publishing Company, Thomson Learning, 1996
Chen X R. Theory of Parameter Estimation in Linear Models (in Chinese). Beijing: Science Press, 1985
Author information
Authors and Affiliations
Corresponding author
Additional information
To commemorate Professor CHEN XiRu (1934–2005)
This work was supported by the President Foundation (Grant No. Y1050) and the Scientific Research Foundation (Grant No. KYQD200502) of GUCAS
Rights and permissions
About this article
Cite this article
Zhang, S., Liao, Y. On some problems of weak consistency of quasi-maximum likelihood estimates in generalized linear models. Sci. China Ser. A-Math. 51, 1287–1296 (2008). https://doi.org/10.1007/s11425-007-0172-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11425-007-0172-7
Keywords
- generalized linear models (GLMs)
- quasi-maximum likelihood estimates (QMLE)
- weak consistency
- convergence rate