Survival Analysis 2

Open the “leuk” data set in STATA (leuk.dta).  The variables in the data set are time (weeks in remission), cens (the indicator of relapse, with 1 if observed relapse and 0 if left study at that time without observed relapse) and group (treatment group; 1 is the 6-MP treatment group and 0 the placebo)

PART A 

1. Produce the plot of the K-M survival estimate and associated pointwise 95% confidence interval for the overall data set.  What are the values of the estimated survival function and its standard error at time 30?  What is the 95% confidence interval for the survival function at time 30?

2. Produce the estimates of the quantiles and their confidence intervals for the placebo group and for the treatment group.  Looking at these values, what do you think about the survival experience of those on the 6-MP treatment compared to the placebo group? 

3. Produce the two different types of estimated means for both the placebo and the 6-MP groups.  Which mean would you use for each group and why?

PART B

4. Produce the plots of the estimated K-M survival functions and the associated confidence intervals for the two groups (placebo and 6MP).  From the plots, do you think this data supports a significant difference in survival experience?

5. Conduct the log-rank test (to include giving appropriate statements for the hypothesis being considered) and give a conclusion.

6. Conduct the Wilcoxon test and again make a conclusion about the two groups.