Consider the time to rearrest data introduced in Section 9.8.
a. Create a Kaplan-Meier curve with a confidence interval for the time until rearrest variable. Describe any patterns you see.
b. Estimate the time at which half the released inmates have been rearrested.
c. Do people with property crimes, person crimes, or both have a longer time before rearrest? You will need to create a new variable to answer this question. Use the new variable to create Kaplan-Meier curves as well as conduct the log- rank test and the Wilcoxon test. Interpret the results.
d. (R software is suggested for this problem.) Plot the estimated hazard function (Kaplan- Meier type) for all the time-to-event data. (Include this graph in your assignment. You can copy and paste R graphs into Word documents.) What do you observe?
(Don’t worry about providing an interpretation. Just give your overall impression of the curve.)
e. (R software is suggested for this problem.) Plot and examine the Nelson- Aalen estimator of the cumulative hazard function. (Include this graph in your assignment.) Discuss the changes in the rate of change of H^(t) as each of the following time periods passes:
• 0 – 3.5 months
• 3.5 – 8 months
• 8 – 11 months
• 11 – 20 months
• 720 months
Also discuss corresponding changes in the estimated hazard as former inmates move from one time interval to another.