Bayesian analysis of partly interval censored data under competing risks framework using engineering data /
Bayesian approach was developed for regression analysis of partly interval-censored data in presence of masking. The primary objective of this thesis is to assess the effect of risks factors (covariates) on both cause-specific hazard (CSH) function and cumulative incident function (CIF), therefore,...
Saved in:
| 主要作者: | |
|---|---|
| 格式: | Thesis |
| 語言: | English |
| 出版: |
Kuala Lumpur :
Kulliyyah of Engineering, International Islamic University Malaysia,
2017
|
| 主題: | |
| 在線閱讀: | http://studentrepo.iium.edu.my/handle/123456789/4368 |
| 標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
| 總結: | Bayesian approach was developed for regression analysis of partly interval-censored data in presence of masking. The primary objective of this thesis is to assess the effect of risks factors (covariates) on both cause-specific hazard (CSH) function and cumulative incident function (CIF), therefore, the popular Cox's proportional hazards model was utilized as well as Fine & Gray's proportional subdistribution hazards model. Several simulations were conducted with various scenarios in order to evaluate the performance of the proposed models. Further, an engineering data set was employed for the purpose of illustration. The proposed methods are flexible and valid for an arbitrary number of causes, moreover, they are easy to implement. The three developed models are shown to have a good performance throughout the simulation studies. The obtained estimations are accurate and comparable to that result from models with only right-censored data and that without masked causes of failure. The model under CIF framework shows a little sensitivity towards high levels of interval-censored observations. Unlike, the models under CSH formulation show sensitivity towards high levels of masking. |
|---|---|
| 實物描述: | xxi, 144 leaves : illustrations ; 30cm. |
| 參考書目: | Includes bibliographical references (leaves 124-128). |