The Korean Journal of Applied Statistics (응용통계연구)

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

A Bayesian zero-inflated Poisson regression model with random effects with application to smoking behavior

랜덤효과를 포함한 영과잉 포아송 회귀모형에 대한 베이지안 추론: 흡연 자료에의 적용

  • 김연경 (중앙대학교 응용통계학과) ;
  • 황범석 (중앙대학교 응용통계학과)
  • Received : 2018.02.13
  • Accepted : 2018.02.17
  • Published : 2018.04.30
Download PDF
⟨ Previous Next ⟩

Abstract

It is common to encounter count data with excess zeros in various research fields such as the social sciences, natural sciences, medical science or engineering. Such count data have been explained mainly by zero-inflated Poisson model and extended models. Zero-inflated count data are also often correlated or clustered, in which random effects should be taken into account in the model. Frequentist approaches have been commonly used to fit such data. However, a Bayesian approach has advantages of prior information, avoidance of asymptotic approximations and practical estimation of the functions of parameters. We consider a Bayesian zero-inflated Poisson regression model with random effects for correlated zero-inflated count data. We conducted simulation studies to check the performance of the proposed model. We also applied the proposed model to smoking behavior data from the Regional Health Survey (2015) of the Korea Centers for disease control and prevention.

0이 과도하게 많이 나타나는 자료는 여러 다양한 분야에서 흔히 볼 수 있다. 이러한 자료들을 분석할 때 대표적으로 영과잉 포아송 모형이 사용된다. 특히 반응변수들 사이에 상관관계가 존재할 때에는 랜덤효과를 영과잉 포아송 모형에 도입해서 분석해야 한다. 이러한 모형은 주로 빈도론자들의 접근방법으로 분석되어왔는데, 최근에는 베이지안 기법을 사용한 분석도 다양하게 발전되어 왔다. 본 논문에서는 반응변수들 사이에 상관관계가 존재하는 경우 랜덤효과가 포함된 영과잉 포아송 회귀모형을 베이지안 추론 방법을 토대로 제안하였다. 이 모형의 적합성을 판단하기 위해 모의 실험을 통해 랜덤효과를 고려하지 않은 모형과 비교 분석하였다. 또한, 실제 지역사회 건강조사 흡연 자료에 직접 응용하여 그 결과를 살펴보았다.

Keywords

References

  1. Angers, J. F. and Biswas, A. (2003). Bayesian analysis of zero-inflated generalized Poisson model, Computational Statistics and Data Analysis, 42, 37-46. https://doi.org/10.1016/S0167-9473(02)00154-8
  2. Celeux, G., Forbes, F., Robert, C. P., and Titterington, D. M. (2006). Deviance information criterion for missing data models, Bayesian Analysis, 1, 651-674. https://doi.org/10.1214/06-BA122
  3. Cohen, A. C. (1963). Estimation in mixtures of discrete distributions. In Proceedings of the International Symposium on Discrete Distributions, Montreal, 373-378.
  4. Gelman, A., Carlin, J. B., Stern, H. S., Dunson, D. B., Vehtari, A., and Rubin, D. B. (2014). Bayesian Data Analysis, CRC Press, New York.
  5. Ghosh, S. K., Mukhopadhyay, P., and Lu, J. C. (2006). Bayesian analysis of zero-inflated regression models, Journal of Statistical Planning and Inference, 136, 1360-1375. https://doi.org/10.1016/j.jspi.2004.10.008
  6. Greene, W. H. (1994). Accounting for excess zeros and sample selection in Poisson and negative binomial regression models, NYU Working Paper, No. EC-94-10.
  7. Haario, H., Saksman, E., and Tamminen, J. (2005). Componentwise adaptation for high dimensional MCMC, Computational Statistics, 20, 265-273. https://doi.org/10.1007/BF02789703
  8. Hall, D. B. (2000). Zero-inflated Poisson and binomial regression with random effects: a case study, Biometrics, 56, 1030-1039. https://doi.org/10.1111/j.0006-341X.2000.01030.x
  9. Jang, H., Kang, Y., Lee, S., and Kim, S. W. (2008). Bayesian analysis for the zero-inflated regression models, The Korean Journal of Applied Statistics, 21, 603-613. https://doi.org/10.5351/KJAS.2008.21.4.603
  10. Lambert, D. (1992). Zero-inflated Poisson regression, with an application to defects in manufacturing, Technometrics, 34, 1-14. https://doi.org/10.2307/1269547
  11. Lee, J., Choi, T., and Woo, Y. (2011a). Bayesian approaches to zero inflated Poisson model, The Korean Journal of Applied Statistics, 24, 677-693. https://doi.org/10.5351/KJAS.2011.24.4.677
  12. Lee, K., Joo, Y., Song, J. J., and Harper, D. W. (2011b). Analysis of zero-inflated clustered count data: a marginalized model approach, Computational Statistics and Data Analysis, 55, 824-837. https://doi.org/10.1016/j.csda.2010.07.005
  13. Li, C. S., Lu, J. C., Park, J., Kim, K., Brinkley, P. A., and Peterson, J. P. (1999). Multivariate zero-inflated Poisson models and their applications, Technometrics, 41, 29-38. https://doi.org/10.1080/00401706.1999.10485593
  14. Liu, H. and Powers, D. A. (2012). Bayesian inference for zero-inflated Poisson regression models, Journal of Statistics: Advances in Theory and Applications, 7, 155-188.
  15. Min, Y. and Agresti, A. (2005). Random effect models for repeated measures of zero-inflated count data, Statistical Modelling, 5, 1-19. https://doi.org/10.1191/1471082X05st084oa
  16. Neelon, B. H., O'Malley, A. J., and Normand, S. L. T. (2010). A Bayesian model for repeated measures zero-inflated count data with application to outpatient psychiatric service use, Statistical Modelling, 10, 421-439. https://doi.org/10.1177/1471082X0901000404
  17. Oh, M. S. and Lim, A. K. (2006). Bayesian analysis of a zero-inflated Poisson regression model: An application to Korean oral hygienic data, The Korean Journal of Applied Statistics, 19, 505-519. https://doi.org/10.5351/KJAS.2006.19.3.505
  18. Ridout, M., Hinde, J., and Demetrio, C. G. B. (2001). A score test for testing a zero-inflated Poisson regression model against zero-inflated negative binomial alternatives, Biometrics, 57, 219-223. https://doi.org/10.1111/j.0006-341X.2001.00219.x
  19. Rodrigues, R. (2003). Bayesian analysis of zero-inflated distributions, Communications in Statistics, 32, 281-289. https://doi.org/10.1081/STA-120018186
  20. Shim, J., Lee, D. H., and Jung, B. C. (2011). Bayesian inference for the zero inflated negative binomial regression model, The Korean Journal of Applied Statistics, 24, 951-961. https://doi.org/10.5351/KJAS.2011.24.5.951
  21. Spiegelhalter, D. J., Best, N. G., Carline, B. P., and Van Der Linde, A. (2002). Bayesian measures of model complexity and fit, Journal of the Royal Statistical Society: Series B, 64, 583-639. https://doi.org/10.1111/1467-9868.00353
  22. Yau, K. K. W. and Lee, A. H. (2001). Zero-inflated Poisson regression with random effects to evaluate an occupational injury prevention program, Statistics in Medicine, 20, 2907-2920. https://doi.org/10.1002/sim.860