This document contains 6 problems related to risk theory and probability distributions. Problem 1 asks to calculate the expected value of a prize amount that follows a mixed Poisson distribution. Problem 2 asks to find the expected value of a random variable with a mixed binomial distribution. Problem 3 asks to calculate the median lifetime for a population with a Weibull distribution and gamma-distributed frailty. The remaining problems involve additional concepts relating to mixed distributions, moment generating functions, and deriving properties of distributions.
This document contains 6 problems related to risk theory and probability distributions. Problem 1 asks to calculate the expected value of a prize amount that follows a mixed Poisson distribution. Problem 2 asks to find the expected value of a random variable with a mixed binomial distribution. Problem 3 asks to calculate the median lifetime for a population with a Weibull distribution and gamma-distributed frailty. The remaining problems involve additional concepts relating to mixed distributions, moment generating functions, and deriving properties of distributions.
This document contains 6 problems related to risk theory and probability distributions. Problem 1 asks to calculate the expected value of a prize amount that follows a mixed Poisson distribution. Problem 2 asks to find the expected value of a random variable with a mixed binomial distribution. Problem 3 asks to calculate the median lifetime for a population with a Weibull distribution and gamma-distributed frailty. The remaining problems involve additional concepts relating to mixed distributions, moment generating functions, and deriving properties of distributions.
This document contains 6 problems related to risk theory and probability distributions. Problem 1 asks to calculate the expected value of a prize amount that follows a mixed Poisson distribution. Problem 2 asks to find the expected value of a random variable with a mixed binomial distribution. Problem 3 asks to calculate the median lifetime for a population with a Weibull distribution and gamma-distributed frailty. The remaining problems involve additional concepts relating to mixed distributions, moment generating functions, and deriving properties of distributions.
13th/14th Feb 2017 1. Bob is a carnival operator of a game in which a player receives a prize worth W = 2N if the player has N successes, N = 0, 1, 2, . . . Bob models the probability of success for a player as follows:
Conditional on = , N has a Poisson distribution with mean
has a uniform distribution on the interval (0, 4)
Calculate E (W ). Hint: for X Poi (), we have MX (t) = e(e 1) . t
2. The random variable N has a mixed distribution
with probability p, N has a binomial distribution with q = 0.5 and m = 2;
with probability 1 p, N has a binomial distribution with q = 0.5 and m = 4;
It is given that Var (N ) = 0.8125. Find E (N ).
3. (Optional) Survival time X for a population of lives follows a Weibull distribution with the following hazard rate function 1 hX| (x|) = x2/3 3 varies over the population with a gamma distribution having parameters = 5, = 0.5. Calculate the median future lifetime for the population. Hint: for X (, ), we have x/ (x/) e ,x > 0 fX (x) = x () 0 , otherwise
MX (t) = (1 t) , t < 1/
4. Fix s 0, and define an Esscher-transformed frailty random variable s with probability density function s e f () , 0 fs () = M (s) 0 , otherwise
(a) Show that s has moment generating function
M (t s) Ms (t) = M (s)
1 (b) Define the cumulant generating function of to be
c (t) = ln [M (t)]
and use (a) to prove that
c0 (s) = E (s ) and c00 (s) = Var (s )
(c) (Optional) For the frailty model, prove that the associated hazard rate may be expressed as
hX (x) = a (x) c0 [A (x)]
where c is defined in (b). Hence, show that
h0 (x) = a0 (x) c0 [A (x)] [a (x)]2 c00 [A (x)]
(d) (Optional) Using (c), prove that if h0X| (x|) 0, then h0X (x) 0.
5. Let N have a Poisson distribution with random mean . Let the distribution of the random variable have probability density function ( 2 ( + 1)1 ( + 1) e , > 0 u () = 0 , otherwise
(a) Obtain the unconditional probability mass function for the mixed Poisson random variable N . (b) i. Find P (z). ii. Hence, find PN (z).
6. [Adapted from STAT3810 Spring 2011 Exam]
Suppose that N | = Poi( + 0.5), and that the MGF of is given by 1 M (t) = . (1 2t)3
Let pn = Pr(N = n), n = 0, 1, 2, . . ., and PN (t) be the probability generating function of N .
(a) Find PN (t).
(b) Find p0 . (c) By considering the derivative of PN (t), derive a recursive formula that expresses pn+1 in terms of pn and pn1 for n = 1, 2, . . .. (d) Find p1 , p2 , p3 and p4 .
