CS2A - Sept23 - EXAM - Clean Proof
CS2A - Sept23 - EXAM - Clean Proof
EXAMINATION
In addition to this paper you should have available the 2002 edition of
the Formulae and Tables and your own electronic calculator.
If you encounter any issues during the examination please contact the Assessment Team on
T. 0044 (0) 1865 268 873.
One of the group decides to estimate the survival of group members using Gompertz
Law with the force of mortality at age x given by:
μ 0.0045 1.0004
(i) Comment on the choice of formula for the force of mortality. [2]
(ii) Calculate the expected number of people that will be at the next dinner stating
any assumptions you make. [5]
(iii) Calculate the expected cost of dinners for this member up to and including
29 February 2040 if each dinner costs £60 per person and ignoring interest
and inflation. [6]
[Total 13]
/
1 𝐶 𝑥 𝐴
𝐹 β 𝑥 α 1 1
𝑛 𝐵
over the range of possible values of the Xn, in the following cases:
(ii) Explain the significance of your results in part (i) by considering the limiting
behaviour of the right-hand side as n → ∞. [8]
[Total 14]
CS2A S2023–2
3 At the start of the week, Alex can choose salad, pizza or sushi for lunch. On Monday
Alex starts with a uniformly random choice, but in the following days makes the
choices following a discrete time Markov chain with transition matrix:
where the 1st, 2nd and 3rd rows and columns correspond to salad, pizza and sushi,
respectively.
(i) Derive the probability that Alex has pizza on Monday and salad on
Wednesday. [4]
(ii) Calculate the probability that Alex has sushi on Wednesday and Friday. [6]
[Total 10]
4 A teacher is looking for ways to cluster her pupils into homogenous groups and
decides to set up an experiment as follows.
Each day, every pupil is given a task and scores ‘+2’ if the task is successfully
completed by the end of the day and ‘−2’ otherwise. This experiment runs for n days
(n > 1) and at the end pupils are grouped based on their total score over n days.
Let us consider a pupil in the class who has a 50% chance of successfully completing
the task each day, and we assume that their performance from day to day is
independent. We denote by Zt this pupil’s score at the end of day t, and by At their
total score from day 1 to day t.
At the end of the experiment, a pupil will be classified as ‘borderline’ if their total
score, An, is zero.
(iii) Determine Pr(A1 = 2 | An = 0) that is the conditional probability that the pupil
scored +2 on the first day given they finish with a borderline classification.
[5]
[Total 11]
CS2A S2023–3
5 Let Xt be the process defined by:
𝑋 𝑌
where:
𝑌 𝑒 𝑏𝑒
(i) State the values of p, d and q for which Xt is an ARIMA(p, d, q) process. [1]
(ii) Demonstrate that var(Yt) = (1 + b2) σ2 and cov(Yt, Yt−1) = bσ2. [4]
var 𝑋 𝑡 1 𝑏 σ 2 𝑡 1 𝑏σ
and that
cov 𝑋 , 𝑋 𝑡 𝑘 1 𝑏 σ 2 𝑡 𝑘 1 𝑏σ
(iv) Explain what the results in part (iii) imply about the shape of the
autocorrelation function of Xt. [2]
[Total 13]
6 A bank branch is responsible for cash replenishments of two cash machines (or
ATMs) near its premises. It is known that the number of customers withdrawing from
the first and the second cash machine follow Poisson distributions at the rate of 20
and 50 per day, respectively.
Withdrawals of cash per customer from the two cash machines follow normal
distributions with mean $1,500 and $1,000 per customer and standard deviation $300
and $200 per customer, respectively.
Assuming that the total cash withdrawals from each of the cash machines
independently follow compound Poisson distributions:
(i) Determine the probability that combined cash withdrawal per customer from
both the cash machines is less than $1,400 per customer. [6]
(ii) Calculate the mean and standard deviation of the combined cash withdrawal
per day from both the cash machines. [8]
[Total 14]
CS2A S2023–4
7 A survival study followed twelve patients, for a maximum of 10 days each, following
a major surgical operation. From previous similar studies, around one-third of patients
survived 10 days. The condition of all patients (i.e. whether a patient was alive or
dead) was monitored daily. The results are set out below where S(t) is the Kaplan–
Meier estimate of the survival function:
(i) Calculate the number of deaths and the number of patients who were censored,
stating the times of all deaths and censoring events. [8]
An expert analyst has voiced concerns about the accuracy of the data, thinking there is
likely to have been an error.
(ii) Briefly explain the likely source of the expert’s concerns. [1]
CS2A S2023–5
8 An insurer has incurred 10,000 claims under a portfolio of home insurance policies.
These claims have a mean size of £2,000 and a standard deviation of £800. One
hundred of these claims have exceeded the excess of loss limit on a reinsurance policy
that the insurer has in place.
(i) Using a Lognormal distribution, estimate the excess of loss limit on this
reinsurance policy. State your assumptions and show all your working clearly.
[5]
(ii) Calculate the number of claims that would be expected to be less than £1,000.
[1]
It has been proposed by the insurance regulator that statutory solvency calculations
should be based on modelling claims using a suitable Normal distribution.
gamma
exponential
Weibull.
[5]
[Total 14]
END OF PAPER
CS2A S2023–6