Scribd
Upload
4 views3 pages

Problem Set 2

The document is a problem set for a course (SI 541) in Spring 2026, consisting of six problems related to survival analysis, hazard functions, and survival functions. It includes tasks such as deriving cumulative hazard functions, survival functions, and analyzing the properties of different survival distributions like exponential and Weibull. The problems require mathematical derivations and interpretations of statistical concepts related to survival times and their implications in population aging.

Uploaded by

nishlesh.hightide
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
4 views3 pages

Problem Set 2

The document is a problem set for a course (SI 541) in Spring 2026, consisting of six problems related to survival analysis, hazard functions, and survival functions. It includes tasks such as deriving cumulative hazard functions, survival functions, and analyzing the properties of different survival distributions like exponential and Weibull. The problems require mathematical derivations and interpretations of statistical concepts related to survival times and their implications in population aging.

Uploaded by

nishlesh.hightide
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Problem Set 2

SI 541
Spring 2026

1. Suppose the hazard function is


(
1, 0 ≤ t < 2,
h(t) =
3, t ≥ 2.

(a) Find the cumulative hazard function H(t).


(b) Derive the survival function S(t).
(c) Find P (T > 3).
(d) Find the density f (t).

2. Let T have survival function S(t).

(a) Starting from


P (t ≤ T < t + ∆t | T ≥ t)
h(t) = lim ,
∆t→0 ∆t
show that
f (t)
h(t) = .
S(t)
(b) Use this result to prove that
 Z t 
S(t) = exp − h(u) du .
0

3. Suppose a fraction p of individuals are immune and will never experience the event,
while the remaining 1 − p have exponential survival with rate λ.

(a) Write the survival function S(t).


(b) Find limt→∞ S(t) and interpret.
(c) Find the hazard function h(t).
(d) Does the hazard converge to 0 or a constant? Explain why.

1
4. Let
1
S(t) = , t ≥ 0.
(1 + t)2

(a) Determine whether the mean survival time


Z ∞
µ= S(t) dt
0

exists or not.
(b) Compute the restricted mean survival time
Z L
µL = S(t) dt
0

for a finite value L > 0.


(c) Interpret why the restricted mean survival time is useful in this setting.

5. Let T be a nonnegative continuous survival time with survival function S(t), hazard
function h(t), and mean residual life (MRL)
R∞
S(u) du
r(t) = E[T − t | T > t] = t .
S(t)

Assume S(0) = 1.
Suppose the mean residual life function satisfies the differential equation

r′ (t) = β − 1 − β h(t) r(t), t > 0,

for some constant β > 0.

(a) Show that the mean residual life and hazard function satisfy the identity

r′ (t) = h(t)r(t) − 1.

(b) Combine the given differential equation with the identity from part (a) to derive
an expression for the hazard function h(t) in terms of r(t) and β.
(c) Solve the resulting differential equation to show that the survival function must
be of the form  −β
t
S(t) = 1 + , t ≥ 0,
θ
for some constant θ > 0.
(d) Identify the probability distribution of T .

2
(e) Determine for which values of β the mean survival time
Z ∞
µ= S(t) dt
0

exists.
(f) Determine whether the hazard function is increasing, decreasing, or constant de-
pending on the value of β. Interpret this in terms of aging properties of the
population.

6. Let T have a Weibull survival function

S(t) = exp(−λtk ), λ > 0, k > 0.

(a) Find the median survival time m.


(b) Show that the mean survival time is
 
1 −1/k 1
µ = E(T ) = λ Γ ,
k k

where Γ(·) is the gamma function.


(c) Compare the mean and median when k > 1. Which is larger? Briefly explain in
terms of skewness of the survival distribution.
(d) Show that  
1 1 −1/k 1 k
r(t) = · λ Γ , λt ,
S(t) k k
where Γ(a, x) is the incomplete gamma function.
(e) Show that r(t) decreases in t when k > 1 and increases when k < 1.
(f) Interpret parts (c) and (f) in terms of aging: explain how the shape parameter k
determines whether the population exhibits aging or “infant mortality.”

You might also like