Lukman, Iing (1999) A Simulation Study on Competing Risks With Censored Data Using Cox Model. Masters thesis, Universiti Putra Malaysia.
A simulation study was performed to compare two regression methods for competing risks with censored data. The first method was the conventional Cox's proportional hazard regression model (Cox model). The second method was based on Cox model using a duplicated data technique of Lunn and McNeil (or the modified Lunn-McNeil). Samples with various sizes and censoring percentages were generated and fitted using both methods. This study was conducted by comparing the inference of both methods, using Root Mean Square Error (RMSE), the power tests, and the Schoenfeld residuals analysis. The power tests used in this study were likelihood ratio test, Rao-score test, and Wald statistics. The Schoenfeld residuals analysis was conducted to check the proportionality of the model through its covariates. The estimated parameters were computed for cause-specific hazards. Results showed the RMSE were generally smaller for the model of the modified Lunn-McNeil method than that of the ordinary Cox method. The power tests of the likelihood ratio statistics and Rao-score test were only powerful for the unstratified Cox model, so that, it could be concluded that the model had more advantages than the modified Lunn-McNeil one. However, results from the analysis of Schoenfeld residuals indicated that the modified Lunn-McNeil was better than the ordinary Cox in complying with the proportional hazards model assumption with respect to certain covariates.
|Item Type:||Thesis (Masters)|
|Subject:||Censored observations (Statistics) - Simulation methods|
|Chairman Supervisor:||Noor Akma Ibrahim, PhD|
|Call Number:||FSAS 1999 3|
|Faculty or Institute:||Faculty of Environmental Studies|
|Deposited By:||Nurul Hayatie Hashim|
|Deposited On:||08 Dec 2010 04:58|
|Last Modified:||11 May 2012 01:15|
Repository Staff Only: Edit item detail
Document Download Statistics
This item has been downloaded for since 08 Dec 2010 04:58.