A Survey on Quasi-Likelihood Estimation Approaches for Longitudinal Set-ups

Authors: Naushad Mamode Khan

Abstract:

The Com-Poisson (CMP) model is one of the most popular discrete generalized linear models (GLMS) that handles both equi-, over- and under-dispersed data. In longitudinal context, an integer-valued autoregressive (INAR(1)) process that incorporates covariate specification has been developed to model longitudinal CMP counts. However, the joint likelihood CMP function is difficult to specify and thus restricts the likelihood-based estimating methodology. The joint generalized quasi-likelihood approach (GQL-I) was instead considered but is rather computationally intensive and may not even estimate the regression effects due to a complex and frequently ill-conditioned covariance structure. This paper proposes a new GQL approach for estimating the regression parameters (GQL-III) that is based on a single score vector representation. The performance of GQL-III is compared with GQL-I and separate marginal GQLs (GQL-II) through some simulation experiments and is proved to yield equally efficient estimates as GQL-I and is far more computationally stable.

Keywords: Longitudinal, Com-Poisson, Ill-conditioned, INAR(1), GLMS, GQL.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1110265

References:

[1] M.A.Al-Osh and A.A.Alzaid, First-order integer-valued autoregressive (INAR(1))process, Journal of Time Series Analysis, 8, 261-275, (1987).
[2] K. Brannas, Explanatory variable in the AR(1) count data model,Umea University, Department of Economics, 381, (1995).
[3] ,J.B. Kadane and G. Shmueli and G. Minka and T.P. Borle and P. Boatwright, Conjugate analysis of the Conway-Maxwell Poisson distribution,Bayesian Analysis, 1, 363-374, (2006).
[4] D. Lord and S.D. Guikema and S. Geedipally, Application of the Conway-Maxwell-Poisson Generalized Linear Model for Analyzing Motor Vehicle Crashes, Accident Analysis and Prevention,40,1123-1134, (2008).
[5] N. Mamode khan and V. Jowaheer, Comparing joint GQL estimation and GMM adaptive estimation in COM-Poisson longitudinal regression model, Communication in Statistics- Simulation and Computation, 42, 755-770, (2013).
[6] P. McCullagh and J.A. Nelder, Generalized Linear Models, Chapman and Hall, (1999).
[7] E. McKenzie, Autoregressive moving-average processes with negative binomial and geometric marginal distrbutions, Advanced Applied Probability, 18, 679-705,(1986).
[8] R.L Prentice and L.P Zhao,Estimating equations for parameters in means and covariances of multivariate discrete and continuous responses, Biometrics, 47,825-839, (1989).
[9] G. Shmueli and T. Minka and J.B. Borle and P. Boatwright, A useful distribution for fitting discrete data, Applied Statistics,Journal of Royal Statistical Society, (2005).
[10] B.C. Sutradhar, An Overview on Regression Models for Discrete Longitudinal Responses, Statistical Science,18, 377-393, (2003).
[11] B.C Sutradhar and K. Das, On the efficiency of regression estimators in generalised linear models for longitudinal data, Biometrika, 86, 459-465, (1999).
[12] B.C. Sutradhar and P. Farrell, Analyzing Multivariate Longitudinal Binary Data: A Generalized Estimating Equations Approach, Canadian Journal of Statistics, 32, 39-55, (2004).
[13] R. Wedderburn, Quasi-likelihood functions, generalized linear models, and the GaussNewton method, Biometrika, 61, 439-447, (1974).