在復發事件資料的分析中，為了考慮不同受試者之間所存在的變異，因此將脆弱項納入風險函數中做考量，以解決事件發生時未觀察到的解釋變項所產生無法觀測之異質性。假設在群體之內的個體擁有相同脆弱項的值，即為共享脆弱模型，其中常見的兩種共享脆弱模型為共享伽瑪脆弱模型或共享對數常態脆弱模型。本研究將以模擬的方法比較兩種模型，最後以躁鬱症患者就診紀錄的復發事件資料做實例分析並運用統計上的三個準則進行兩模型比較。

In the analysis of recurrent events data, mostly consider the variability among different subjects. Therefore, include frailty term into the hazard function to measure heterogeneity caused by unobserved covariates. It means individuals in the cluster share the same frailty term, thus the model is called shared frailty model. More common models are shared gamma and log-normal frailty models which we compare with each other in terms of the simulation. Finally, we separately apply two models to medical records of patients with bipolar disorder and make comparisons by three statistical criterions.

Akaike, H. (1998). Information theory and an extension of the maximum likelihood principle Selected Papers of Hirotugu Akaike (pp. 199-213): Springer.
Andersen, P. K., & Gill, R. D. (1982). Cox's regression model for counting processes: a large sample study. The annals of statistics, 1100-1120.
Clayton, D. G. (1978). A model for association in bivariate life tables and its application in epidemiological studies of familial tendency in chronic disease incidence. Biometrika, 65(1), 141-151.
Cox, D. R. (1992). Regression models and life-tables Breakthroughs in statistics (pp. 527-541): Springer.
Duchateau, L., & Janssen, P. (2008). the frailty model (Statistics for Biology and Health). Springer: New York, 10, 978-970.
Erhan, C. (1975). inlar. Introduction to stochastic processes: Prentice-Hall.
Hougaard, P. (2012). Analysis of multivariate survival data: Springer Science & Business Media.
Joly, P., Commenges, D., & Letenneur, L. (1998). A penalized likelihood approach for arbitrarily censored and truncated data: application to age-specific incidence of dementia. Biometrics, 185-194.
Katz, R. W. (1981). On some criteria for estimating the order of a Markov chain. Technometrics, 23(3), 243-249.
Klein, J. P. (1992). Semiparametric estimation of random effects using the Cox model based on the EM algorithm. Biometrics, 795-806.
Lewis, P. A., & Shedler, G. S. (1979). Simulation of nonhomogeneous Poisson processes by thinning. Naval Research Logistics (NRL), 26(3), 403-413.
Liang, K.-Y., Self, S. G., Bandeen-Roche, K. J., & Zeger, S. L. (1995). Some recent developments for regression analysis of multivariate failure time data. Lifetime data analysis, 1(4), 403-415.
Marquardt, D. W. (1963). An algorithm for least-squares estimation of nonlinear parameters. Journal of the society for Industrial and Applied Mathematics, 11(2), 431-441.
McGilchrist, C., & Aisbett, C. (1991). Regression with frailty in survival analysis. Biometrics, 461-466.
Rondeau, V., Commenges, D., & Joly, P. (2003). Maximum penalized likelihood estimation in a gamma-frailty model. Lifetime data analysis, 9(2), 139-153.
Rondeau, V., Mazroui, Y., & Gonzalez, J. R. (2012). frailtypack: an R package for the analysis of correlated survival data with frailty models using penalized likelihood estimation or parametrical estimation. J Stat Softw, 47(4), 1-28.
Schwarz, G. (1978). Estimating the dimension of a model. The annals of statistics, 6(2), 461-464.
Shibata, R. (1976). Selection of the order of an autoregressive model by Akaike's information criterion. Biometrika, 63(1), 117-126.
Sullivan, J. J. (1988). Three roles of language in motivation theory. Academy of Management Review, 13(1), 104-115.
Vaupel, J. W., Manton, K. G., & Stallard, E. (1979). The impact of heterogeneity in individual frailty on the dynamics of mortality. Demography, 16(3), 439-454.
Wienke, A. (2010). Frailty models in survival analysis: CRC Press.