Here is the code for the non-parametric procedure survival and covariates:
PROC LIFETEST Data=Survival_Analysis Plots=(h,ls,lls) Atrisk Outs=Survival_Table; Time tenure*censor(1); ID custid; Test Risk_appetite Fund_performance Inv_potential Inv_involvement AUM Complex_prod Complaints; Run;
This produces the negative log graph, among others:
In Figure 6.10, the negative log of the estimated survivor function passes through the origin and looks a fit case of exponential distribution:
The cumulative hazard is rising exponentially over time. In the ...