簡易檢索 / 詳目顯示
| 研究生： |
陳和謙 Ho-Chien Chen |
|---|---|
| 論文名稱： |
區間設限下的存活模型預測準確度 Survival model based predictive accuracy for interval censored data |
| 指導教授： |
曾議寬
Yi-Kuan Tseng |
| 口試委員: | |
| 學位類別： |
碩士 Master |
| 系所名稱： |
理學院 - 統計研究所 Graduate Institute of Statistics |
| 論文出版年： | 2020 |
| 畢業學年度： | 108 |
| 語文別： | 中文 |
| 論文頁數： | 234 |
| 中文關鍵詞： | AFT加速失效模型 、一致性指標 、Cox比例風險模型 、區間設限 |
| 外文關鍵詞： | accelerated failure time model, Concordance index, Cox proportional hazards model, interval censored data |
| 相關次數： | 點閱：17 下載：0 |
| 分享至: |
| 查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
一致性指標（Concordance index ,又稱為C-index）經常被用來衡量模型的預測準確性，此指標能運用在未設限與設限資料下。在過去文獻中，可以使用參數模型或無母數方法來獲得區間設限資料下的一致性指標。此研究希望將回歸模型引入一致性指標中，使得模型的預測準確度更加準確，因此利用Cox模型的資訊來得到一致性指標。而當比例風險假設不成立時，在這種情況下則可以使用AFT加速失效模型作為替代。但區間設限中半母數模型的估計是困難的 ，在Cox模型下，本文引用Anderson-Bergman (2015)所提出的演算法，來估計區間設限中半母數模型參數。此外當使用AFT模型時，Gao, Zeng和Lin (2017)提出的方法。最後在模擬研究來觀察一致性指標的表現，並將該方法應用於HIV資料。
The Concordance index(C-index) was often used to measure the prediction accuracy of the model. This index can be used under uncensored and censored data. In the literature, we can use the parametric model or nonparametric method to obtain the C-index under the interval censored data. In this study, we would like to introduce the hazard regression models into the C-index to make the model prediction accuracy more efficient. The Cox model is used to incorporate into C-index due to its flexibility. Moreover, in case that proportional assumption fails, the accelerated failure time (AFT) model was used to replace the Cox model. The estimation of the semi-parametric model in interval censored data is not straighforward. Under the Cox model, we use the algorithm proposed by Anderson-Bergman (2015) to estimate the regression parameters in interval censored data. In addition, we use the method proposed by Gao, F., Zeng, D.,and Lin, D. Y. (2017) when AFT model is used. The performance of C-index is accessed by simulation study and the proposed method is applied to HIV data.
[1]Anderson-Bergman, C. (2016). An efficient implementation of the EMICM algorithm for the interval censored NPMLE. Journal of Computational and Graphical Statistics. 81(15), 1-24.
[2]Anderson-Bergman, C. (2017). icenReg: Regression Models for Interval Censored Data in R. Journal of Computational and Graphical Statistics 81(12), 1-23.
[3]Gao, F., Zeng, D., and Lin, D. Y. (2017). Semiparametric Estimation of the Accelerated Failure Time Model with Partly Interval-Censored Data. Biometrics 73, 1161–1168.
[4]Goggins, W. B. and Finkelstein, D. M. (2000). A proportional hazards model for multivariate interval-censored failure time data. Biometrics 56, 940–943.
[5]Gonen, M. and Heller, G. (2005). Concordance probability and discriminatory power in proportional hazards regression. Biometrika 92, 965–970.
[6]Graf, E., Schmoor, C., Sauerbrei, W., and Schumacher, M. (1999). Assessment and comparison of prognostic classification schemes for survival data. Statistics in Medicine 18(17-18), 2529–2545.
[7]Harrell, F. E., Lee, K. L., and Mark, D. B. (1996). Multivariate prognostic models: Issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Statistics in Medicine 15, 361–387.
[8]Heagerty, P. J. and Zheng, Y. (2005). Survival model predictive accuracy and roc curves. Biometrics 61, 92-105.
[9]Huang, J., Wellner, J. (1997).Interval Censored Survival Data: A Review of Recent Progress.In Proceedings of the First Seattle Symposium in Biostatistics: Survival Analysis, 123–169. Springer-Verlag, New York.
[10]Liu, X., Jin, Z., and Graziano, J. H. (2012). Comparing paired biomarkers in predicting quantitative health outcome subject to random censoring. Statistical Methods in Medical Research 25, 447–457.
[11]Tsouprou, S. (2015). Measures of discrimination and predictive accuracy for interval censored survival data. Mathematical Institute Master Thesis, Leiden University Medical Center, Nederland.
