REGRESI COX DENGAN ESTIMASI TERBOBOTI MENGGUNAKAN METODE KAPLAN - MEIER; COX REGRESSION WITH WEIGHTED ESTIMATION USING KAPLAN – MEIER METHOD

Simohatono, Arif M, Danardono

2015 | Skripsi | FMIPA

Cox regression is a well-known approach for modeling censored survival data. However, the model has an assumption of proportional hazard which requires an attention. In certain cases, non-proportional hazard assumption is frequently violated i.e. in the long-term study. In this studi, we examine the Cox regression with weitghted estimation model under non-proportional hazard. This model using maximum partial likelihood for estimating the parameter. The weighting function is obtained from survival function using Kaplan-Meier method. Using Cox regression with weighted estimation, we can estimate ? parameter from non-proportional hazard covariate, which cannot be obtained with stratified Cox regression. The proposed and existing model of Cox regression with weighted estimation are applied to a data set in heroin addict. We will show the estimate of the parameter in the data set.

Kata Kunci : survival analisis; Cox regression; weighted estimation