Efficient estimation in ZIP models with applications to count data
Keywords:
Quadratic estimating functions, Linear estimating functions, Count data, Zero-inflatedAbstract
Estimating functions have been used in estimating parameters of many continuous time series
models. However, this method has not been applied to models involving count data. In
this paper, we use quadratic estimating functions (QEF) to derive estimators for the joint estimation of the conditional mean and variance parameters of count data models, specifically
the basic zero-inflated Poisson (ZIP) model, ZIP regression model and integer-valued generalized autoregressive heteroscedastic model with ZIP conditional distribution. Results show that the estimators derived from QEF method, which uses information from combined estimating functions, is more informative than linear estimating functions (LEF) method that only uses information from component estimating functions. Finally, we also fit the real data sets using the ZIP models via QEF, LEF and maximum likelihood methods, and in so doing, demonstrate the superiority of the QEF method in practice.
References
Abaza, B. (1982). A recursive formulation for
the two-step least squares estimator. Kuwait
Journal of Science, 9: 77-85.
Allen, D., Ng, K.H. & Shelton, P. (2013).
Estimating and simulating Weibull models
of risk or price durations: An application to
ACD models. North American Journal of Economics
and Finance, 25: 214-225.
Bahadur, R.R. (1958). Examples of inconsistency
of maximum likelihood estimates.
Sankhya: The Indian Journal of Statistics, 20:
-210.
Baksh, M.F., Bhning, D. & Lerdsuwansri,
R. (2011). An extension of an over-dispersion
test for count data. Computational Statistics
and Data Analysis, 55(1): 466-474.
Bera, A.K., Bilias, Y. & Simlai, P. (2006).
Estimating functions and equations: An essay
on historical developments with applications
to econometrics. Palgrave Handbook of
Econometrics, 1: 427-476.
Brockwell, P.J. & Davis, R.A. (1991). Time
series: Theory and methods. Springer Science
& Business Media. New York. pp. 106-110.
Chan, K.S. & Ledolter, J. (1995). Monte
Carlo EM estimation for time series models
involving counts. Journal of the American Statistical Association, 90(429): 242-252.
Chandra, S.A. & Taniguchi, M. (2001). Estimating
functions for nonlinear time series
models. Annals of the Institute of Statistical
Mathematics, 53(1): 591-597.
Crowder, M. (1987). On linear and quadratic
estimating functions. Biometrica, 74(3): 591-
Godambe, V.P. (1985). The foundations of finite
sample estimation in stochastic processes.
Biometrika, 72(2): 419-428.
Godambe, V.P. & Heyde, C.C. (2010).
Quasi-likelihood and optimal estimation. Selected
Works of CC Heyde, Springer. New
York. pp. 386-399.
Kharrati-Kopaei, M. & Faghih, H. (2011).
Inferences for the inflation parameter in the
ZIP distributions: The method of moments.
Statistical Methodology, 8(4): 377-388.
Lambert, D. (1992). Zero-inflated Poisson
regression, with an application to defects in
manufacturing. Technometrics, 34(1): 1-14.
Liang, Y., Thavaneswaran, A. & Abraham,
B. (2011). Joint estimation using quadratic estimating
function. Journal of Probability and
Statistics. DOI:10.1155/2011/372512.
Merkouris, T. (2007). Transform martingale
estimating functions. The Annals of Statistics,
(5): 1975-2000.
Nanjundan, G. & Naika, T.R. (2012).
Asymptotic comparison of method of moments
estimators and maximum likelihood estimators
of parameters in zero-inflated Poisson
model. Applied Mathematics, 3(6): 610-616.
Ng, K.H. & Peiris, S. (2013). Modelling high
frequency transaction data in financial economics:
A comparative study based on simulations.
Economic Computation and Economic
Cybernetics Studies and Research, 47(2): 189-
Ng, K.H., Shelton, P., Thavaneswaran, A. &
Ng, K.H. (2015). Modelling the risk or price
durations in financial markets: Quadratic estimating
functions and applications. Economic
Computation and Economic Cybernetics Studies
and Research, 49(1): 223-238.
Staub, K.E. &Winkelmann, R. (2012). Consistent
estimation of zero-inflated count models.
Health Economics, 22(6): 673-686.
Thavaneswaran, A. & Abraham, B. (1988).
Estimation for nonlinear time series models
using estimating equations. Journal of Time
Series Analysis, 9(1): 673-686.
Thavaneswaran, A., Ravishanker, N. &
Liang, Y. (2015). Generalized duration models
and optimal estimation using estimating
functions. Annals of the Institute of Statistical
Mathematics, 67(1): 129-156.
Vinod, H. (1997). Using Godambe-Durbin
estimating functions in econometrics. Lecture
Notes-Monograph Series. pp. 215-237.
(http://www.jstor.org/stable/4356018)
Zhu, F. (2012). Zero-inflated Poisson and negative
binomial integer-valued GARCH models.
Journal of Statistical Planning and Inference,
(4): 826-839.
