過去在研究地中海果蠅產卵量與壽命之間的關係，通常是依照總產卵量大小對果蠅進行分組，來觀察產卵數與壽命之間的相關性，本研究是以果蠅產卵量因子分成接近連續的方式，將1000隻雌性地中海果蠅產卵量由小至多排序，以累積樣本的方式來觀察產卵量對壽命的影響有什麼樣的變化。在探討時間相依共變數與存活時間之間的關係時，一般較常使用的半母數方法是以部分概似法來估計參數，前提是必須具有完整的共變數資訊及沒有測量誤差，才能夠達到準確地估計。為了減少使用部分概似法對參數的偏誤，可以使用聯合模型來做配適，以Cox比例風險模型與加速失敗模型的風險函數作為架構，主要目的是比較部分概似法與聯合模型對共變數與時間有關之下，其參數估計值的變化與影響，以動態的曲線來探討總產卵量的多寡是否影響參數的估計以及變數與存活時間之間的關係，並尋找果蠅的樣本會造成每日產卵量對存活的效應在Cox模型和AFT模型中有矛盾的結論。

In the literature, Mediterranean fruit fly is usually classified into a few discrete groups based on the total numbers of eggs laid to observe the relationship of the numbers of eggs and the life expectancy. In our thesis, we adopt semiparametric method for approximately continuous groups according to the total eggs laid, We are interested in the change of impact of daily egg laying when sample size increasing. The semiparametric methods applied including Cox partial likelihood method and full likelihood joint model method. For partial likelihood approach, the complete covariate information and zero measuring error are required, so that we can estimate accurately. If we consider the time-dependent covariates measured with random errors, joint model then is employed with survival described by Cox proportional hazards model or accelerated failure time model. Dynamic curve of regression coefficients as a function of total egg-laying are plotted to identify the change of impact of daily egg-laying on survival. Moreover, from the dynamic curves, we identify specific samples may have contradict conclusions using Cox model or AFT model.

[1]Carey, J. R. & Liedo, P. (1995). Life Table Aging Rates in Large Medfly Cohorts. Experimental Gerontology, 30, 315-325.
[2]Carey, J. R. & Liedo, P. (1995). Sex Mortality Differentials and Selective Survival in Large Medfly Cohorts. The Gerontologist, 35, 588-596.
[3]Carey, J. R., Liedo, P. & Vaupel, J. W. (1995). A Male-Female Longevity Paradox in Medfly Cohorts. Journal of Animal Ecology, 64, 107-116.
[4]Carey, J. R. & Liedo, P. (1995). Mortality Dynamics of Density in the Mediterranean Fruit Fly. Experimental Gerontology, 30, 605-629.
[5]Carey, J. R. (1997). What Demographers Can Learn from Fruit Fly Actuarial Models and Biology. Demography, 34, 17-30.
[6]Carey, J. R., Liedo, P., M\"uller, H. G.,Wang, J. L. & Chiou, J. M. (1998). Relationship of Age Patterns of Fecundity to Mortality, Longevity, and Lifetime Reproduction in a Large Cohort of Mediterranean Fruit Fly Females. Journal of Cenmtologv: Biological Sciences, 53A, B245-B251.
[7]Ciampi, A. & Etezadi-Amoli, J. (1985). A general model for testing the proportional hazards and the accelerated failure time hypothesis in the analysis of censored survival data with covariate. Communications in Statistics, 14, 651-667.
[8]Cox, D. R. (1972). Regression Models and Life-Tables. Journal of the Royal Statistical Society. Series B, 34, 187-220.
[9]Cox, D. R. & Oakes, D. (1984). Analysis of Survival Data. London:Chapman and Hall.
[10]Dafini, U. G. & Tsiatis, A. A. (1998). Evaluating surrogate markers of clinical outcome measured with error. Biometrics, 54, 1445-1462.
[11]Dempster, A. P., Laird, N. M. & Rubin, D. B. (1977). Maximum Likelihood from Imcomplete Data via the EM Algorithm. Journal of the Royal Statistical Society. Series B, 39, 1-38.
[12]Ding, J. & Wang, J. L. (2008). Modeling Longitudinal Data with Nonparametric Multiplicative Random Effects Jointly with Survival Data. Biometrics, 64, 546-556.
[13]Efron, B. & Tibshirani, R. J. (1993). An introduction to the Bootstrap. Chapman and Hall, New York.
[14]Henderson, R., Diggle, P. & Dobson, A. (2000). Joint Modeling of Longitudinal Measurements and Event Time Data. Biometrics, 1, 465-480.
[15]Hsieh, F.,Tseng, Y. K. & Wang, J. L. (2006). Joint Modeling of Survival and Longitudinal Data:Likelihood Approach Revisited. Biometrics, 62, 1037-1043.
[16]Rizopoulos, D., Verbeke, G. & Molenberghs, G. (2008). Shared Parameter Models under Random Effects Misspecification. Biometrika, 95, 63-74.
[17]Rizopoulos, D., Verbeke, G. & Lesaffre, E. (2009). Fully Exponential Laplace Approximations for the Joint Modeling of Survival and Longitudinal Data. Journalof the Royal Statistical Society B, 71, 637-654.
[18]Song, X., Davidian, M. & Tsiatis, A. (2002). A Semiparametric Likelihood Approach to Joint Modeling of Longitudinal and Time-to-Event Data. Biometrics, 58, 742-753.
[19]Tseng, Y. K., Hsieh, F. & Wang, J. L. (2005). Joint Modeling of Accelerated Failure Time and Longitudinal Data. Biometrics, 92, 587–603.
[20]Tseng, Y. K., Su, Y. R., Mao, M. & Wang, J. L. (2015). An Extended Hazard Model with Longitudinal Covariates. Biometrika, 102, 135–150.
[21]Tsiatis, A. A. & Davidian, M. (2004). Joint Modeling of Longitudinal and Time-to-Event Data: an overview. Statistica Sinica, 14, 809–834.
[22]Wulfsohn, M. S. & Tsiatis, A. A. (1997). A Joint Model for Survival and Longitudinal Data Measured with Error. Biometrics, 53, 330-339.
[23]Zeng, D. & Cai, J. (2005). Asymptotic results for maximum likelihood estimators in joint analysis of repeated measurements and survival time. The Annals of Statistics, 33, 2132–2163.
[24]高欣如 (2006)。Cox 比例風險假設之探討與擴充風險模型之應用。國立中央大學統計研究所碩士論文。
[25]陳婉婷 (2006)。Cox 比例風險模型之參數估計-比較部分概似法與聯合模型。國立中央大學統計研究所碩士論文。
[26]黃穎慈 (2005)。以聯合模型探討地中海果蠅繁殖力與老化的關係。國立中央大學統計研究所碩士論文。