We consider the leukemia data from Exercise 3.1. a) Compute the Kaplan-Meier estimates for the 6-MP group and for the placebo group. b) Plot both Kaplan-Meier estimates in the same figure. What can...

We consider the leukemia data from Exercise 3.1.

a) Compute the Kaplan-Meier estimates for the 6-MP group and for the placebo group.

b) Plot both Kaplan-Meier estimates in the same figure. What can you learn from the plots?

In Section 3.2.6 it is proved that, for given value of t, the Kaplan-Meier estimator S(t) is approximately normally distributed with mean S(t) and a variance that may be estimated by τ 2 (t) given by (3.27) [or alternatively by Greenwood’s formula (3.28)]. In Exercise 3.3 we discussed (in the context of the Nelson-Aalen estimator) how one may derive confidence intervals based on transformations. Use this approach to show that the transformation g(x) = −log(−logx) yields the logminus-log transformed confidence interval (3.30) for the survival function.