Statistical comparison of survival models between cox model and exponential model for analysis of breast cancer patients
Elsamoual Mohammed Kurtokeila, Dr. Hamza Ibrahim Hamza, Dr. Amal Alsir Alkhidir
This paper discuss the comparison between semi parametric model (Cox regression) and parametric model (Exponential model) according to survival time distribution. To evaluate the factors that effect on the survival of the breast cancer patients, to determine the best model Akaike Information Criterion (AIC) and Bayesian information criterion (BIC) was used. A sample of 246 was taken from Khartoum Oncology Hospital, 150 (61%) died. The results of data analysis showed that the Hazard Ratios for the factors studied in Exponential Model is lower than the Hazard Ratios for the factors studied in Cox Model. AIC and BIC criterion in Exponential model is less than Cox Model, this indicate that the Exponential model is a favourable for the data of this study.