The problem is to fit the Cox model that fits the best in R. Please, check out the file attached. It has a Cox model already, but if you check the plot of residuals it is clear that this model doesn't fit well. The problem is to build the best fitted model step by step, or using BIC, or by doing transformations if necessary, or include interaction terms. All the work needs to be done in R withing 60 hours after the winner is selected. Thank you.