Scribd
Upload
17 views

assignment4 (2)

Uploaded by

aniketgupta.techfest
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
17 views

assignment4 (2)

Uploaded by

aniketgupta.techfest
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
You are on page 1/ 3

EE 325: Probability and Random Processes

Homework on Statistics and Hypothesis Testing

1. The material for this is from parts of Sections 8.1-8.4 of the Papoulis textbook. The
book has a lot more material but I have covered 8.1, most of 8.2, the MLE in Section
8.3, and some of the material before testing of distributions in Section 8.4.

2. There are several solved examples in the text. As usual, you are expected to go
through them and understand them.

3. Solve problems 8.3–8.7, 8.11, 8.12, 8.21, and 8.22.

4. Estimation is also well covered in the Hajek text; see Sections 2.8, 3.7, and 4.9 for a
basic understanding

1
EE 325: Probability and Random Processes
Programming Assignment 4: Part a
To be completed before 30 October 2024

In this assignment we will determine the premiums for a life insurance policy. The insurance
scheme works as follows. Customers can buy a one-year insurance policy that covers them
for the insured amount, say Ω. Each insured person will pay a premium, say α, that is
determined by the amount insured and the age of the person. If there is a death of the
insured the person in the year, then the family is paid Ω otherwise there is no payout to
the insurer from the company. Note that the number of people that will buy the policy in
a year and the number of people for whom the insured amount has to be paid are random
variables.

1. Let X denote the number of insurers in a year. Assume that this is a random variable.
Let Y be the number of claims in the year. We will assume that X is a Poisson random
variable with mean λ and each of policy will result in a claim with probability p.
Determine the joint pmf of X and Y. The surplus, denoted by S, at the end of the
year is (αX − ΩY) . Clearly, the surplus is a random variable that is a function of the
joint pmf of X and Y and the premium is set so that the surplus being less than some
a is a low probability event. Typically, the company begins the year with a sufficient
endowment and some loss, i.e., a negative surplus, may also be allowed. For a given
Ω, assume you know λ, and p, the objective is to fix the premium α such that the
probability of ‘breaking the pot’ is sufficiently low.
Before performing calculations, determine the marginal pmf of Y.
Assume Ω = 10. We will consider two values of λ—10,000 and 100,000, and two
values of p—0.01 and 0.02. For a = 10, 0, −10, −20, −30, −100, find the minimum α
such that Pr(S < 0.01) . Tabulate your results for the four combinations of (λ, p) and
the six values of a.
Rather than using ratios of factorials of large numbers, you can use Gaussian approx-
imation suitably in the probability formula that you calculated above. Specifically,
note that Gaussian approximations for the Poisson and the binomial distributions
are available; the latter was done in class. These can perhaps be used suitably in the
formulas that you have derived to compute the joint pmf.

2. Let us take the same situation as above. except that there are two types of policies
targeted at two different demographics with (Ωi , αi ) targeting a population demo-
graphic characteristic of (λi , pi ,) for i = 1, 2. Here the insurance company is willing to
do some cross subsidy. Here we can consider three surpluses—the surpluses from each
type of policy, denoted by Si and the total surplus, S and want that Pr(Si < ai ) < 0.01

2
for I = 1, 2, and Pr(S < a) < 0.01. Determine α1 and α2 for the following parameters:
λ1 = 100, 000, λ2 = 5, 000, Ω1 = 10, Ω2 = 100 p1 = 0.02, p2 = 0.02, a1 = −100,
a2 = −15 and a = 105.

3. Be ready with the programs for the above two parts before 29 October. In Part b of
this assignment, you will be given data for N years for the number of policies sold
and the number of claims in each year. You will have to estimate λ and p and then
use these estimates suitably to determine the α.

Discussion

• In both the parts you are essentially searching for the ‘right α.0 And you will need
to perform summations over a large number of indexes. Using the numerical insights
discussed in class, and developing more of those insights on your own, you can suitably
reduce the search space.

• For the second part above several values of α1 , α2 would be possible. Which of those
would you choose?

• What do you think ”cross subsidy” means? How would you achieve it? Argue for
and against cross subsidy in this case.

• If, instead of using the joint pmf of X and Y above, you had assumed that they were
independent would your answers for the two parts change substantially. Explain.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy