Estimation of Count Data using Bivariate Negative Binomial Regression Models

نوع مقاله: مقاله پژوهشی

نویسندگان

1 هیات علمی دانشگاه آزاد

2 عضو هیئت علمی دانشگاه رازی کرمانشاه

3 استاد دانشگاه ukm

4 دانشجوی دکتری اقتصاد دانشگاه upm

چکیده

Abstract
Negative binomial regression model (NBR) is a popular approach for modeling overdispersed count data with covariates. Several parameterizations have been performed for NBR, and the two well-known models, negative binomial-1 regression model (NBR-1) and negative binomial-2 regression model (NBR-2), have been applied. Another parameterization of NBR is negative binomial-P regression model (NBR-P), which has an additional parameter and the ability to nest both NBR-1 and NBR-2. This paper introduces several forms of bivariate negative binomial regression model (BNBR) which can be fitted to bivariate count data with covariates. The main advantages of having several forms of BNBR are that they are nested and allow likelihood ratio test to be performed for choosing the best model, they have flexible forms of mean-variance relationship, they can be fitted to bivariate count data with positive, zero or negative correlations, and they allow overdispersion of the two dependent variables. Applications of several forms of BNBR are illustrated on two sets of count data; Australian health care and Malaysian motor insurance.

کلیدواژه‌ها


عنوان مقاله [English]

Estimation of Count Data using Bivariate Negative Binomial Regression Models

نویسندگان [English]

  • Pouya Faroughi 1
  • Mohaamad Sharif Karimi 2
  • Ismail Noriszura 3
  • Asrin Karimi 4
1 Assistant Professor, (Corresponding Author), Department of Statistics, Sanandaj Branch, Islamic Azad University, Sanandaj, Iran.
2 Assistant Professor, Department of Economics, Faculty of Social Sciences, Razi University, Kermanshah, Iran.
3 Professor, School of Mathematical Sciences, Faculty of Science and Tchnology, Universiti Kebangsaan Malaysia
4 Faculty of Economics and Management, University Putra Malaysia
چکیده [English]

Abstract
Negative binomial regression model (NBR) is a popular approach for modeling overdispersed count data with covariates. Several parameterizations have been performed for NBR, and the two well-known models, negative binomial-1 regression model (NBR-1) and negative binomial-2 regression model (NBR-2), have been applied. Another parameterization of NBR is negative binomial-P regression model (NBR-P), which has an additional parameter and the ability to nest both NBR-1 and NBR-2. This paper introduces several forms of bivariate negative binomial regression model (BNBR) which can be fitted to bivariate count data with covariates. The main advantages of having several forms of BNBR are that they are nested and allow likelihood ratio test to be performed for choosing the best model, they have flexible forms of mean-variance relationship, they can be fitted to bivariate count data with positive, zero or negative correlations, and they allow overdispersion of the two dependent variables. Applications of several forms of BNBR are illustrated on two sets of count data; Australian health care and Malaysian motor insurance.

کلیدواژه‌ها [English]

  • bivariate Poisson regression model
  • bivariate negative binomial regression model
  • Correlation
  • overdispersion
Berkhout, P. & E.Plug. (2004). A bivariate Poisson Count Data Model Using Conditional Probabilities. Stat Neerl 58:349-364. doi: 10.1111/j.1467-9574.2004.00126.x.## Cameron, A.C. & P. Johansson. (1997). Count Data Regression Using Series Expansions: With Applications. J Appl Econom 12:203-223. Doi: 10.1002/ (SICI) 1099-1255(199705)12:33.0.CO;2-2##Cameron, A.C., T. Li, P.K. Trivedi & D.M. Zimmer. (2004). Modelling the Differences in Counted Outcomes Using Bivariate Copula Models With Application to Mismeasured Counts. The Econometrics Journal 7: 566-584. doi: 10.1111/j.1368-423X.2004.00144.x. ##Cameron, A.C. & P.K. Trivedi. (1986) Econometric Models Based on Count Data: Comparisons and Applications of Some Estimators and Tests. J Appl Econom 1:29-53. doi: 10.1002/jae.3950010104. ##Cameron, A.C. & P.K. Trivedi. (2013). Regression Analysis of Count Data. Cambridge University Press, New York. ##Cameron, A.C., P.K. Trivedi, F. Milne & J. Piggott. (1988). A Microeconomic Model of the Demand for Health Care and Health Insurance in Australia. Rev Econ Stud, 55:85-106. ##Campbel, J.T. (1934). The Poisson Correlation Action. Proceedings of the Edinburg Mathematical Society 4:18-26. ##Chernoff, H. (1954). On the Distribution of The Likelihood ratio. Ann Math Stat 25:573-578. ##Famoye, F. (2010). On the Bivariate Negative Binomial Regression Model. J Appl Stat 37: 969-981. Doi: 10.1080/02664760902984618. ##Famoye, F. (2012). Comparisons of Some Bivariate Regression Models. J Stat Comput Simul. 82(7): 937-949. ##Faroughi, P. & N. Ismail. (2014). A New Bivariate Negative Binomial Regression Model.  International Conference on Quantitative Sciences and Its Applications (Icoqsia 2014): Proceedings of the 3rd International Conference on Quantitative Sciences and Its Applications. Vol. 1635. No. 1. AIP Publishing. ##Gourieroux, C., C. Monfort & A. Trognon. (1984). Pseudo Maximum Likelihood Methods: Applications to Poisson Models. Econometrica 52:701-720. ##Greene, W. (2008). Functional Forms for the Negative Binomial Model for Count Data. Econ Lett 99: 585-590. doi:10.1016/j.econlet.2007.10.015. ##Gurmu, S. & J. Elder. (2000). Generalized Bivariate Count Data Regression Models. Econ Lett 68:31-36. ##Hilbe, J. (2011). Negative Binomial Regression. Cambridge University Press, Cambridge.  ##Johnson, N., S. Kotz & N. Balakrishnan. (1997). Discrete Multivariate Distributions, John Wiley and Sons, Inc: New York. ##Jung, R.C. & R. Winkelmann. (1993). Two Aspects of Labor Mobility: A Bivariate Poisson Regression Approach. Empi Econ 18:543-556. doi: 10.1007/BF01176203. ##Karimi, A., P. Faroughi & K.A. Rahim. (2015). Modeling and Forecasting of International Tourism Demand in ASEAN Countries. Am J Appl Sci, 12(7): 479- 486. doi: 10.3844/ajassp.2015.479.486. ##King, G. (1989). A Seemingly Unrelated Poisson Regression Model. Sociol Method Res 17:235-255. doi: 10.1080/1350485032000082018B. ##Kocherlakota. S. & K. Kocherlakota. (2001). Regression in the Bivariate Poisson Distribution. Commun Stat Theor M 30:815-825. doi: 10.1081/STA-100002259. ##Lakshminarayana, J., S.N.N. Pandit & K.S. Rao. (1999). On A Bivariate Poisson Distribution. Commun Stat Theor M 28:267-276. doi: 10.1080/03610929908832297. ##Lawless, J.F. (1987). Negative Binomial and Mixed Poisson Regression. Can J Stat 15:209-225. doi: 10.2307/3314912. ##Lee, A. (1999). Modelling Rugby League Data Via Bivariate Negative Binomial Regression.  Aust Zn J Stat: 41:141-152. doi: 10.1111/1467-842X.00070. ##Mitchell, C.R. & A.S. Paulson. (1981). A New Bivariate Negative Binomial Distribution. Nav Res Logist Q 28:359–374. doi: 10.1002/nav.3800280302. ##Ophem, H.V. (1999). A general method to estimate correlated discrete random variables. Economet Theor 15:228-237, 1999. doi: 10.1017/S0266466699152058. ##Ridout, M.S, J.P. Hinde & C.G.B. Demetrio. (2001). A Score Test for Testing A Zero-Inflated Poisson Regression Model Against Zero-Inflated Negative Binomial Alternatives. Biometrics 57:219-223. doi: 10.1111/j.0006-341X.2001.00219.x. ##Self, S.G. & K. Liang. (1987). Asymptotic Properties of Maximum Likelihood Estimators and Likelihood Ratio Tests Under Nonstandard Conditions. J Am Stat Assoc 82:605-610. doi: 10.1080/01621459.1987.10478472. ##Winkelmann, R. (2008). Econometric Analysis of Count Data. Springer Verlag, Heidelberg. ##Zamani, H. & N. Ismail. (2012). Functional form for the Generalized Poisson Regression Model. Commun Stat Theor M 41:3666-3675. doi: 10.1080/03610926.2011.564742. ##Zamani, H. & N. Ismail. (2013). Score Test for Testing Zero-Inflated Poisson Regression Against Zero-Inflated Generalized Poisson Alternatives. J Appl Stat 40:2056-2068. doi: 10.1080/02664763.2013.804904. ##Zulkifli, M., N. Ismail & A.M. Razali. (2013). Analysis of Vehicle Theft: A Case Study in Malaysia Using Functional Forms of Negative Binomial Regression Models. Appl Math Inform Sci 7: 389-395. dx.doi.org/10.12785/amis/072L02. ##