Abstarct: With the advancement of technology in data storage, recently the big data are appeared in clinical trials. Determining the effective variables in predicting the time of death in a disease leads to save cost and also time, and is therefore of great importance. Nowadays in many medical studies, microscopic technology is used to find the relationship between the behavior of the genes and the various clinical features of the patients. In this dissertation, the three famous penalty function, ridge, LASSO and a combination of these two function, Elastic net are used in a Cox proportional hazards regression model and compared.