•
~~'-ii1
UNIVERSITY OF GHANA
(All rights reserved)
BSC/BA. FIRST SEMESTER EXAMINATIONS: 2018/2019
DEPARTMENT OF STATISTICS & ACTUARIAL SCIENCE
STAT 331: PROBABILITY DISTRIBUTIONS (3 CREDITS)
INSTRUCTION: Answer all questions in Section A and two questions from
Section B.
TIME ALLOWED: THREE (3) HOURS
Section A (50 marks)
I Answer all questions I
AI. Suppose that a production company produces number of non-defective items (X) which
is distributed as binomial with parameters Nand fJ, where 0 < fJ < 1. If the random
variable N has the Poisson distribution with parameter A, show that the moment
[Link] function of X is given by
Mx{t) = exp{AfJ(et - I)}.
[6 Marks]
A2. Suppose F is a O'-algebra of subsets of 0 and let B E F.
(a) Show that {A n 13 : A E F} is a O'-algebra of subsets of B. [4 marks]
(b) If the sum of possible outcomes of a fair die thrown twice is wand the sample
space 0 = {2: 3, 4, 5, 6, 7, 8, 9,10,11, l2}, and let B = {2, 4, 6, 8, 10} E F, then
deduce that for any subset A of 0, {A n B : A E F} is a O'-algebra of subsets of
B. [5 marks]
A3. Given that a random variable X have a Poisson distribution with A;
(a) Find E [cos{7l'X)] if E(X) = ln2. [6 marks]
Examiners: Dr. Louis Asiedu/ Mr. Gabriel Kallah-Dagadu Page 1 of 6
(b) Suppose A is a. random variable with probability density function given by;
where n is a fixed positive constant. Show that;
1( k + n) ( 1 ) k+n
f(x) = 1(n)r(k + 1) 2 '
k = 0,1,2, ...
{
0, otherwise
hence find the distribution of f(x), if n is an integer or \:In E Z. [6 marks]
A4. X has a discrete Uniform distribution on the integers 0,1,2,3, ... , nand Y has a
discrete uniform distribution on the integers 1,2,3, ... , n. Show that
Vo,r[X]- Vo,r[Y] = (2n+ 1)
12
[6 marks]
A5. (a) Let Tl be the time of the first event in Poisson process {X(t), t > O} with
E[X(t)] = At, A> O. Find the P(T1 < 3IX(5) = 1). [4 marks]
(b) Suppose a random variable X has a Poisson distribution with parameter Y. If
Y is also a Poisson random variable with parameter (), find the factorial moment
generating function (f.m.g.f) of Z = X + Y. [4 marks]
A6. Consider X the number of successes in x + m Bernoulli trials each with a success
probability Y. Suppose that Y is a random variable with uniform distribution over the
interval (0,1). Find and identify P(X = x) . [4 marks]
A7. Suppose that the random variable W has beta distribution with probability density
function f(w) given a..<;
f(w) = Kw~(1- w)~, 0 < w < 1.
Find [Link] value of K and hence the average value of W. [5 marks]
Examiners: Dr. Louis Asiedu/ Mr. [Link] Kallah-Dagadu Page 2 of 6
Section B (50 marks)
Answer any two questions from this section.
All questions carry equal marks.
B1. (a) Let {Yn ; n = 1,2,3: ... } be a sequence of random variables. Explain what it
means to say that
1. Yn. has a limiting distribution as n ---t 00. [3 marks]
11. the limiting distribution of Yn is degenerate. [3 marks]
(b) The random variable Xn. has probability density function
ne-n.(x-6) x > 0
fn.(x) = { 0: otherwise
Using moment generating functions, show that
i. Xn has a limiting distribution which is degenerate at x = e. [5 marks]
n. Zn = n(Xn - e) has a limiting distribution which is exponential with mean
1. [6 marks]
(c) Given that Yi: i = 1: 2,3: ... ,n are independent and identically distributed
random variable with E(Yi) = Ii and V(Yi) = (j < 00, Show that the distribution
of Un,
2:~=1 Yi - nf.L
u - n. - (jVn
converges to the standard normal distribution function as n ---t 00. [8 marks]
Hint: Let Y1 , Y 2, Y 3 ... be a sequence of random variables having moment
generating functions m( t), ml (t), m2(t), m3 (t): . .. respectively. If
limn-+oomn(t) = m(t)
then the distribution function of Y" converges to the distribution function of Y.
Examiners: Dr. Louis Asiedu/ Mr. Gabriel Kallah-Dagadu Page 3 of 6
B2. (a) Suppose some events occur independently of each other at a steady mean rate A
per unit time in such a way that if Pn(t) denotes the probability that n events
occur in the interval (0, t], then Pn(t) satisfies the following conditions;
A. For small h, Pn(h) = Ah + o(h);
B. For small h, L.m=2 Pn(h) = o(h);
C. The numbers of events in non-overlapping intervals of time are stochastically
independent and have the same distribution for the same length of interval;
D. poCO) = 1
Prove that,
(Atte->'t
Pn(t) = n.
I ,n = 0,1,2, ...
[8 marks]
Deduce that;
(b) i. If Tn denotes the occurrence time of the nth event, then Tn has a gamma
distribution with parameters n and A. [5 marks]
11. Suppose phone calls are assumed to arrive at the department of Statistics and
Actuarial Science office according to a Poisson process at a mean of 30 per
hour. Find the probability that five calls arrive in fifteen minutes interval.
[3 marks]
(c) In any Poisson process {N(t), t 2: O}, for s < t then the
P{N(s) = kIN(t) = n}= ( nk ) (S)k
t (1 - ts)n-k . [4 marks]
(d) If N1(t), t 2: 0 and N 2 (t), t 2: 0 are independent Poisson processes, with respective
rates or intensities VI and V2, then for 0 ::; k ::; n,
1
where P = ----,--,----,- [5 marks]
1 + (vdvr)
Examiners: Dr. Louis Asieduj Mr. Gabriel Kallah-Dagadu Page 4 of 6
B3. (a) Suppose that Xi; i = 1,2, ... , n are independent random variables with a common
uniform distribution over (0, 1) and Y k is uniformly distributed over (0, Yk-d with
Yo = 1, k = 1,2, ... , n. Let
k
Zk = II Xi, k = 1,2,3, ... ,n.
i=l
Show by mathematical induction or any other method that Zk has probability
density function
(-In z)k-l
fk(Z) = (k _ I)!' k = 1,2, ... ,n.
Hence deduce that Yk and Zk have the same distribution (k = 1,2, ... ,n).
[9 marks]
(b) Suppose that N, Xl, X 2 , ... , X N are independent non-negative integer-valued
random variables. Let R = Xl + X 2 + ... + X N , the sum of random variables
with a random number of terms (N is random) and suppose that Xis have the
same p.m.f with common p.g.f , g(s).
1. gR(S) = gN[g(S)], where gR(S) is p.g.f of Rand gN(S) is p.g.f of N.
[3 marks]
ii. E(R) = E(N)E(X). [2 marks]
iii. VaT(R) = [E(X)J2VaT(N) + E(N)VaT(X) [3 marks]
(c) Let X be the number of defective products produced by a manufacturing company
for every N number of products in a batch. Suppose that the each product
produced has a probability e of being defective independent of other products.
Find the probability distribution of X if the total number of products N in a
batch is a random variable with probability mass function given by;
P(N = k) = { (~)ek(l - e)M-k k = 0, 1,2, ... ,M
0, otherwise
[8 marks]
Examiners: Dr. Louis Asiedu/ Mr. Gabriel Kallah-Dagadu Page 5 of 6
B4. (a) Suppose that the number of omissions of a-particles from two different factories
processing radioactive materials within a specified time interval t is represented
by X and Y respectively. If the random variables X and Yare distributed as
Poisson with parameters At and [Link] respectively, and the random variable Z =
X + Y represent the total number of a-particles from the two factories, find the
probability mass function of Z. [6 marks]
(b) In a process of sintering (heating) two types of copper powder, the density function
for Y1 , the volume proportion of solid copper in a sample, was given by
f( )= { 6Yl (1 - Yl), 0:::; Yl :::; 1
Yl 0, elsewhere
The density function for Y 2 , the proportion of type A crystals among the solid
copper, was given as
3y~, 0:::; Y2 :::; 1
g(Y2) = { 0, elsewhere
The variable U = Y1 Y2 gives the proportion of the sample volume due to type A
crystals. If Y1 and Y2 are independent, find the probability density function for
U.
[7 marks]
(c) Suppose X and Yare independent random variables with the common uniform
distribution over (0,1) and Zl is uniformly distributed over (0, X). If Z2 = XY,
show that Z2 has a density function
f (z) = -In z, °< z < 1.
Deduce that Zl and Z2 have the same distribution. [5 marks]
(d) A random variable X has cumulative density function
F
(x) =
{ 1 - e-~:2 , x > °
0, otherwise
Find all differential monotonic function X = w(z) such that z has a uniform
distribution on the interval [0,1]. [7 marks]
Examiners: Dr. Louis Asiedu/ Mr. Gabriel Kallah-Dagadu Page 6 of 6