Scribd
Upload
64 views

Tutorial 2

This document contains 7 questions related to survival analysis and the Kaplan-Meier estimator, Nelson-Aalen estimator, and Cox proportional hazards model. It provides various datasets on times until events such as battery failure, organ transplant rejection, candle burn times, insurance policy termination, medical trial treatment times, and health insurance claim times. For each question, it asks the reader to calculate estimates of the survival function, distribution function, or partial likelihood using the given statistical methods and data.

Uploaded by

Najia Syukur
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
64 views

Tutorial 2

This document contains 7 questions related to survival analysis and the Kaplan-Meier estimator, Nelson-Aalen estimator, and Cox proportional hazards model. It provides various datasets on times until events such as battery failure, organ transplant rejection, candle burn times, insurance policy termination, medical trial treatment times, and health insurance claim times. For each question, it asks the reader to calculate estimates of the survival function, distribution function, or partial likelihood using the given statistical methods and data.

Uploaded by

Najia Syukur
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

SMA4053 Actuarial Model | Yumn S.

Yusoff

Tutorial 2.
1. My camera uses expensive batteries and I have been collecting data to see
how long they last. However, if I am going on a trip, I replace the battery
before I go. The number of rolls that I shot on each battery is (with * denoting a
battery replaced prior to a trip)
17, 13, 15*, 7*, 21, 18*, 5, 18, 6*, 22, 19*, 15, 4, 11, 14*, 18, 10, 10, 8*, 17
Construct a table which lists the Kaplan-Meier estimator at each time of
battery failure.

2. A clinical trial has been conducted in order to assess the effectiveness of a


new drug in preventing rejection of an organ transplant. Thirty-six patients who
received an organ transplant have been observed over a one-year period.
Eighteen of these patients were treated with the drug (the treatment group)
and eighteen received a placebo (the control group). The times (in weeks) until
rejection or censoring (without rejection) for each group is shown below. Note
that the individuals received the transplants at different times over the year, so
the follow-up time for each individual is different.

Treatment Group
Times to rejection: 6, 6, 7, 10, 13, 16, 23
Times to censoring: 6, 9, 11, 12, 19, 20, 25, 30, 32, 32, 35

Control Group
Times to rejection: 1, 1, 2, 3, 5, 5, 8, 9, 9, 10, 10, 11, 12, 18, 25
Times to censoring: 3, 4, 9

(a) Calculate the Kaplan-Meier estimate of the survival function for both
groups.
(b) Calculate the Kaplan-Meier estimate of the distribution function, 𝐹̂ (𝑡) for
both groups.

3. Refer the data in Question 2, calculate the Nelson Aalen estimate of survival
function and the distribution function, 𝐹̂ (𝑡).

4. The following data give the burn time of a number of candles of the same type.
During the experiment some candles went out due to other reasons than the
wax being fully usurped and hence there are censored values which are
denoted by a *.
Time (hours): 5*, 6*, 6.5, 6.5, 6.5*, 7, 8*, 8.5, 9, 9, 9, 9.5.
Use the data to calculate, the Kaplan-Meier estimator of the survival function
of the burn time.
5. The following data represents the time at which a five-year term insurance policy
terminates for 30 policyholders. The cause of termination for each policyholder
is either death, surrender or expiration of the five-year period. There are twelve
policies left at the end of the expiration period.

Policy Event Death/surrender Policy Event Death/


time surrender
time
1 s 0.1 11 d 2.9
2 d 4.8 12 s 3.9
3 s 0.8 13 d 4.0
4 d 0.8 14 s 4.0
5 d 3.1 15 s 4.1
6 s 1.8 16 d 4.8
7 s 2.1 17 s 4.8
8 s 2.5 18 s 4.8
9 s 2.8 19-30 e 5.0
10 d 2.9

“s” indicates surrender, “d” indicates death, “e” indicates expiration of the 5-
year period. Calculate the Nelson Aalen estimator of the survival function.

6. The table gives the data from a small medical trial on the time (in hours) until
monitoring indicates that further treatment is necessary:

Drug A 3 6+ 8+ 10
Drug B 5 7+ 9+ 12
where the + indicates that the observation was censored.
(i) Show the partial likelihood in the Cox proportional hazards model for
comparing the two drugs.
(ii) Let 𝑙 = 𝑙(𝛽) = log 𝐿 (𝛽). Show that 𝛽̂ = −0.792

7. The table gives the data from a small portfolio of permanent health insurance
(PHI) policies, and shows, for policyholders classified by sex, (Male, Female),
the time in months until a claim is made; the + indicates that the observation
was censored.

Male 1 2+ 3+ 7 11 13+ 15 20
Female 2 4+ 6+ 8 12+ 16 23
Suppose that the Cox Proportional Hazards Model holds for these data and
that male is taken as the reference category, find the partial log likelihood 𝑙(𝛽)
𝑑𝑙
and deduce an expression for 𝑑𝛽.

You might also like