2007 STAT3955 Past Paper 1

.pdf

School

The University of Hong Kong *

*We aren’t endorsed by this school

Course

3811

Subject

Statistics

Date

Nov 24, 2024

Type

pdf

Pages

5

Uploaded by KidNeutron12719

THE UNIVERSITY OF HONG KONG DEPARTMENT OF STATISTICS AND ACTUARIAL SCIENCE STAT3811 SURVIVAL ANALYSIS May 22, 2007 Time: 2:30 p.m. - 5:30 p.m. Candidates taking examinations that permit the use of calculators may use any cal- culator which fulfils the following criteria: (a) it should be self-contained, silent, battery-operated and pocket-sized and {b) it should have numeral-display facilities only and should be used only for the purposes of calculation. It is the candidate's responsibility to ensure that the calculator operates satisfactorily and the candidate must record the name and type of the calculator on the front page of the examination scripts. Lists of permitted/prohibited calculators will not be made available to candidates for reference, and the onus will be on the candidate to ensure that the calculator used will not be in violation of the criteria listed above. Answer ALL SIX questions. Marks are shown in square brackets. 1. The following data consist of the times to relapse and death of 10 bone marrow transplant patients. Table 1 Patient Relapse Time (months) Death time (month) 1 5 11 2 8 12 3 12 15 4 24 33+ 5 32 45 6 17 28+ 7 16+ 16+ 8 17+ 17+ 9 19+ 19+ 10 30+ 30+ In the sample, patients 4 and 6 were still alive at the end of the study and patients 7-10 were alive, free of relapse at the end of the study. Suppose the time to relapse has an exponential distribution with hazard rate A and the time to death has a Weibull distribution with parameters () and a.
S&AS: STAT3811 Survival Analysis 2 (a) Construct the likelihood function for the relapse rate ,\ and calculate the MLE of A. (b) Construct the likelihood function for the parameters B and a. (c) Suppose we were only allowed to observe a patient's death time if the patient relapsed. Construct the likelihood for B and a based on this truncated sample, and compare it to the results in (b). [Total: 15 marks] 2. Consider the data on the time to relapse of 10 bone marrow transplant patients in Table 1 above. 3. (i) Construct the Kaplan-Meier Product-Limit (PL) estimator and plot the estimated survival curve. (ii) Calculate the estimated variance of the PL estimator and determine the 95% confidence interval for the survival probability S(t) at t = 20. (iii) Calculate the mean remission survival time to 30 months. (iv) Find the median remission survival time. [Total: 15 marks] The following data consist of the relapse survival time (in days) of twenty patients after treatment for a rare type of cancer 20 48 68 106+ 119+ 129 132 134+ 142 198+ 201 208+ 215 231+ 246 290+ 302 304+ (a) Estimate the cumulative hazard rate using the Nelson-Aalen estimator. (b) Estimate the hazard rate at t = 80 days, using the uniform kernel-with a bandwidth of 40 days. (c) Estimate the hazard rate at t = 250 days, using the uniform kernel with a bandwidth of 40 days. Hint: The uniform kernel is given by K(x) = 1/2 for -1 ~ x ~ 1, 0 otherwise. [Total: 15 marks] 4. Ten female patients with breast cancer are randomized to receive either CMF (cyclic administration of cyclophosphamide, methatrexate, and fluorouracil) or no treatment after a radical mastectomy. At the end of two years, the fol- lowing times to relapse (or remission times) in months are recorded: CMF (group 1) 23, 16+, 18+, 20+, 24+ Control (group 2) 15, 18, 19, 19, 20 184+ 318
S&AS: STAT3811 Survival Analysis 3 (a) Suppose the significance level a= 0.05. Use the Mantel-Haenszel test to test the null hypothesis Ho: S1(t) = S 2 (t) two treatments are equally effective against H 1 : 8 1 (t) =/:- S 2 (t) two treatments are not equally effective (b) Assume that the two distributions are exponential with parameters .X 1 and A2, respectively. Under the same significance level as (a), use the Cox F- test to test the null hypothesis Ho : .X 1 = .X 2 two treatments are equally effective against H 1 : .X 1 =/:- .X 2 two treatments are not equally effective [Total: 15 marks] 5. In a study of noise level and efficiency, twelve students were given a very simple test under three different noise levels. It is known that under normal conditions, they should be able to finish the test in 10 minutes. The students were randomly assigned to the three levels. Table 2 gives the time required to finish the test for the three levels. Are the three noise levels significantly different at the 0.05 level of significance? (See the statistical table on P.5) Table 2 1 2 3 10.5 10.0 12.0 9.0 12.0 13.0 9.5 12.5 15.5 9.0 11.0 14.0 [Total: 15 marks]
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help