DOI QR코드

DOI QR Code

영과잉 공간자료의 분석

Zero In ated Poisson Model for Spatial Data

  • 한준희 (양산부산대학교병원 연구통계지원실) ;
  • 김창훈 (부산대학교 의학전문대학원)
  • Han, Junhee (Research And Statistical Support, Research Institute of Convergence for Biomedical Science and Technology, Pusan National University Yangsan Hospital) ;
  • Kim, Changhoon (Department of Preventive Medicine, Pusan National University School of Medicine)
  • 투고 : 2015.03.16
  • 심사 : 2015.03.31
  • 발행 : 2015.04.30

초록

가산자료(counts data)를 적합 하는 경우 보통 포아송 모형이 가장 먼저 고려된다. 과산포 문제가 있을 경우도 유사 포아송(quasi Poisson) 모형이나 음이항(Negative binomial) 모형으로 대부분 설명이 가능하다. 하지만, 가산자료 중에는 포아송분포를 가정한 기대 빈도 이상으로 많은 0이 관측되는 자료가 있고 이를 영과잉(Zero inflated) 가산 자료라고 부른다. 영과잉 가산자료를 설명하기 위해 영과잉 포아송(ZIP) 모형이나 영과잉 음이항(ZINB) 모형을 이용할 수 있다. 더 나아가 영과잉 가산자료가 공간상관관계까지 있을 경우 영과잉 문제뿐만 아니라 유의할 수 있는 공간효과까지 고려해야하고 이를 위해 혼합효과모형(mixed effects model)이 고려 될 수 있다. 본 연구에서 사용된 2004년 기준 부산시 남성동별 갑상선암 발생자수 자료를 이용하여, 일반선형 포아송모형, 영과잉 포아송모형, 공간 영과잉 포아송모형을 적합하여 비교해보았다.

A Poisson model is the first choice for counts data. Quasi Poisson or negative binomial models are usually used in cases of over (or under) dispersed data. However, these models might be unsuitable if the data consist of excessive number of zeros (zero inflated data). For zero inflated counts data, Zero Inflated Poisson (ZIP) or Zero Inflated Negative Binomial (ZINB) models are recommended to address the issue. In this paper, we further considered a situation where zero inflated data are spatially correlated. A mixed effect model with random effects that account for spatial autocorrelation is used to fit the data.

키워드

참고문헌

  1. Kim, D. J., Ki, M., Kim, M. H., Kim, Y. M., Yoon, T. H., Jang, S. R., JangChoi, K. H., Kang, A. R., Chae, H. R. and Choi, J. H. (2014). Developing Health Inequalities Indicators and Monitoring the Status of Health Inequalities in Korea, 2014-04, Korea Institute for Health and Social Affairs.
  2. Ahn, H. S., Kim, H. J. and Welch, H. G. (2014). Korea's thyroid-cancer "epidemic"-screening and overdiagnosis, The New England Journal of Medicine, 371, 1765-1767. https://doi.org/10.1056/NEJMp1409841
  3. Chen, A. Y., Jemal, A. and Ward, E. M. (2009). Increasing incidence of differentiated thyroid cancer in the United States, 1988-2005, Cancer, 115, 3801-3807. https://doi.org/10.1002/cncr.24416
  4. Cho, B. Y., Choi, H. S., Park, Y. J., Lim, J. A., Ahn, H. Y., Lee, E. K., Kim, K. W., Yi, K. H., Chung, J. K., Youn, Y. K., Cho, N. H., Park do, J. and Koh, C. S. (2013). Changes in the clinicopathological characteristics and outcomes of thyroid cancer in Korea over the past four decades. Thyroid : official journal of the American Thyroid Association, 23, 797-804. https://doi.org/10.1089/thy.2012.0329
  5. Ferlay, J., Shin, H. R., Bray, F., Forman, D., Mathers, C. and Parkin, D. M. (2010). Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008, International Journal of Cancer. Journal International du Cancer, 127, 2893-2917. https://doi.org/10.1002/ijc.25516
  6. Han, M. A., Choi, K. S., Lee, H. Y., Kim, Y., Jun, J. K. and Park, E. C. (2011). Current status of thyroid cancer screening in Korea: results from a nationwide interview survey, Asian Pacific Journal of Cancer Prevention: APJCP, 12, 1657-1663.
  7. Jung, K. W., Won, Y. J., Kong, H. J., Oh, C. M., Cho, H., Lee, D. H. and Lee, K. H. (2015). Cancer statistics in Korea: Incidence, mortality, survival, and prevalence in 2012. Cancer Research and Treatment : Official Journal of Korean Cancer Association.
  8. Kim, D. J., Ki, M., Kim, M. H., Kim, Y. M., Yoon, T. H., Jang, S. R., JangChoi, K. H., Kang, A. R., Chae, H. R. and Choi, J. H. (2014). Developing Health Inequalities Indicators and Monitoring the Status of Health Inequalities in Korea, 2014-04, Korea Institute for Health and Social Affairs.
  9. Kweon, S. S., Shin, M. H., Chung, I. J., Kim, Y. J. and Choi, J. S. (2013). Thyroid cancer is the most common cancer in women, based on the data from population-based cancer registries, South Korea. Japanese Journal of Clinical Oncology, 43, 1039-1046. https://doi.org/10.1093/jjco/hyt102
  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. H. and Shin, S. W. (2014). Overdiagnosis and screening for thyroid cancer in Korea, Lancet, 384, 1848. https://doi.org/10.1016/S0140-6736(14)60984-3
  12. Lee, T. J., Kim, S., Cho, H. J. and Lee, J. H. (2012). The incidence of thyroid cancer is affected by the characteristics of a healthcare system. Journal of Korean Medical Science, 27, 1491-1498. https://doi.org/10.3346/jkms.2012.27.12.1491
  13. Li, N., Du, X. L., Reitzel, L. R., Xu, L. and Sturgis, E. M. (2013). Impact of enhanced detection on the increase in thyroid cancer incidence in the United States: review of incidence trends by socioeconomic status within the surveillance, epidemiology, and end results registry, 1980-2008. Thyroid : Official Journal of the American Thyroid Association, 23, 103-110. https://doi.org/10.1089/thy.2012.0392
  14. Londero, S. C., Krogdahl, A., Bastholt, L., Overgaard, J., Pedersen, H. B., Frisch, T., Bentzen, J., Pedersen, P. U., Christiansen, P. and Godballe, C. (2013). Papillary thyroid carcinoma in Denmark 1996-2008: An investigation of changes in incidence, Cancer Epidemiology, 37, e1-6. https://doi.org/10.1016/j.canep.2012.10.011
  15. Moran, P. A. P. (1950). Notes on continuous stochastic phenomena, Biometrika, 37, 17-23. https://doi.org/10.1093/biomet/37.1-2.17
  16. Spiegelhalter, D. J., Best, N. G., Carlin, B. P. and van der Linde, A. (2002). Bayesian measures of model complexity and fit (with discussion), Journal of the Royal Statistical Society, Series B, 64, 583-639. https://doi.org/10.1111/1467-9868.00353
  17. The BUGS project j MRC Biostatistics Unit, WinBUGS 1.4.3, (2007). www.mrc-bsu.cam.ac.uk/software/bugs/.