fit cox proportional hazard regression model by choosing a set of 3 to 4 covariates (one covariate must be categorical with two labels. for example, race and tpr variables for class practice data and hw1 data could be categorical variable with two labels) and a dependent variable (time to event variable: