Assignment Part I

Statistics for quantitative finance

Autumn 2015

• This Assignment Part I is individual, containing seven questions.

• Three freely chosen questions are compulsory, the remaining are supplementary.

• Each compulsory question yields 10 marks.

• The solution of the supplementary questions gives additional marks up to a sum of 30 marks.

• The submission date is Thursday 30/04/2015 before the lecture at 6:00pm.

• The weight of the Assignment Part I is 20%.

QUESTION 1 (10 marks)

At the station there are three pay phones which accept 20 cent pieces. One never works, another
always works, while the third works with probability 1/2. On my way to the metropolis for the
day, I wish to identify the reliable phone, so that I can use it on my return. The station is empty
and I have just three 20 cent pieces. I try one phone and it does not work. I try another twice in
succession and it works both times. What is the probability that this second phone is the reliable
one?

QUESTION 2 (10 marks)

You have n urns, the r -th of which contains r − 1 red balls and n − r blue balls, r = 1 . . . , n. You
pick an urn at random and remove two balls from it without replacement. Find the probability
that the two balls are of different colors.

QUESTION 3 (10 marks)

An expedition is sent to the Himalayas with the objective of catching a pair of wild yaks for
breeding. Assume yaks are loners and roam about the Himalayas at random. The probability
p ∈ [0, 1[ that a given trapped yak is male is independent of prior outcomes. Let N be the
random variable which stands for the number of yaks that must be caught until a pair is obtained.
Calculate the distribution of N and determine E(N ) depending on p.

QUESTION 4 (10 marks)

A coin shows heads with probability p ∈]0, 1[ and tails with probability q = 1 − p. Let Xn be
the number of tosses needed to obtain n heads. Using Tschebysheff’s inequality, estimate the
probability P(X200 > 400) from above.

This study source was downloaded by 100000816115074 from CourseHero.com on 05-13-2023 08:44:53 GMT -05:00

https://www.coursehero.com/file/83506358/Assignments-2015pdf/
 QUESTION 5 (10 marks)
Let X be a random variable, uniformly distributed on [0, 1]. Consider its transformation

a + bX
Y = , a, b, c, d > 0 with d < c
c − dX
Calculate the density of Y .

QUESTION 6 (10 marks)

Assume that the random variables X1 and X2 are independent, and each follows an exponential
distribution with parameter λ > 0. Determine the joint distribution of

Y1 = X1 + X2 , Y2 = X1 /X2

QUESTION 7 (10 marks)

Suppose that (Xn )n∈N are independent identically distributed random variables each following
uniform distribution on [0, 1]. Apply Central Limit Theorem to determine the limit
−1 1
lim P((X1 · X2 · · · · · Xn )n 2
en 2 ∈ [a, b]), a, b ∈ R+ .
n→∞

13

This study source was downloaded by 100000816115074 from CourseHero.com on 05-13-2023 08:44:53 GMT -05:00

https://www.coursehero.com/file/83506358/Assignments-2015pdf/
 Solutions assignment Part I
Statistics for quantitative finance
Autumn 2015

Solution 1 (10 marks)

Let A be the event in the question: the first phone tried did not work and second worked
twice. Clearly:

P(A | 1st reliable) = 0

further
1 1
P(A | 2d reliable) = P(1st never works | 2d reliable) + P(1st works half | 2st reliable)
2 2
1 1 1 3
= + × =
2 2 2 4
finally
1 1 1
P(A | 3d reliable) = × × P(2d works half | 3d reliable) =
2 2 8
That is, the required probability is
3/4 × 1/3 6
P(2d reliable) = =
0 × 1/3 + 3/4 × 1/3 + 1/8 × 1/3 7

Solution 2 (10 marks) We have

n
! 1
P(1st red, 2nd blue) = P(1st red, 2nd blue | urn k chosen) ×
n
k=1
n
1 ! (k − 1)(n − k) 1
= = ··· =
n (n − 1)(n − 2) 6
k=1
"n
which is obtained using formula i=1 i(i − 1) = 13 (n + 1)n(n − 1). Finally,

P(1st blue, 2nd red) = P(1st red, 2nd blue), P(different colors) = 1/3

Solution 3 (10 marks)

We have P(N = n) = pn−1 (1 − p) + (1 − p)n−1 p giving the expectation
∞
! ∞
! ∞
!
E(N ) = = (1 − p) npn−1 + p n(1 − p)n−1
n=2 n=2 n=2

we have
∞
! ∂
(1−p)× npn−1 = (1−p)× |p=0 (1/(1−p)−1−p) = (1−p)×(1/(1−p)2 −1) = 1/(1−p)−(1−p)
n=2
∂p

similarly
∞
!
p× n(1 − p)n−1 = 1/p − p
n=2

This study source was downloaded by 100000816115074 from CourseHero.com on 05-13-2023 08:44:53 GMT -05:00

https://www.coursehero.com/file/83506358/Assignments-2015pdf/
 thus
1 1 1−p p
E(N ) = + +1= + +1
1−p p p 1−p
Solution 4 (10 marks) Since X200 follows negative binomial distribution, with r = 200 and
success probability p. Thus we have
n n(1 − p)
E(X) = , Var(X) = .
p p2
giving
2 )
E(X200 E(X200 )2 + Var(X200 )
P(X200 ≥ 400) ≤ 2
=
400 4002
2
200 + 200(1 − p)
≤
4002 p2
which is a reasonable estimate for large p only.
Solution 5 (10 marks) Define the transformations
a a+b a + bx
h :]0, 1[→] , [, x '→
c c−d c − dx
with the inverse
a a+b cy − a
h−1 :] , [→]0, 1[, h−1 (x) = .
c c−d b + dy
The random variable Y = h(X) possess the density
a a+b
fY (y) = fX (h−1 (y))|(h−1 )% (y)|, y ∈] , [.
c c−d
With the density fX = 1[0,1] we obtain
ad + bc
fY (y) = 1] a , a+b [ (y)|(h−1 )% (y)| = 1] a , a+b [ (y) , y ∈ R.
c c−d c c−d (b + dy)2
Solution 6 (10 marks) Consider the map
x1
h : R2+ :→ R2+ , (x1 , x2 ) '→ (x1 + x2 , )
x2
with the inverse
y1 y2 y1
h−1 : R2+ :→ R2+ , (y1 , y2 ) '→ ( , ).
1 + y2 1 + y2
and Jacobian # $
y1 y1 y1 y2
1+y2 1+y2 −
(1+y2 )2
J(y1 , y2 ) = 1 y1
.
1+y2 − (1+y 2)
2

giving the determinant

y1
| det(J(y1 , y2 ))| = | |, (y1 , y2 ) ∈ R2+
(1 + y2 )2
The joint density is given by

fY1 ,Y2 (y1 , y2 ) = fX1 ,X2 (h−1 (y1 , y2 ))| det(J(y1 , y2 ))|
y1 y2 y1 y1
= fX1 ,X2 ( , )| |
1 + y2 1 + y2 (1 + y2 )2
y y
−λ 1 2 −λ 1
y
y1
= λe 1+y2 λe 1+y2 | |
(1 + y2 )2
1
= λ2 y1 e−λy1
(1 + y2 )2

15

This study source was downloaded by 100000816115074 from CourseHero.com on 05-13-2023 08:44:53 GMT -05:00

https://www.coursehero.com/file/83506358/Assignments-2015pdf/
 for all (y1 , y2 ) ∈ R2+ showing that Y1 and Y2 are independent, with Y1 and Y2 following the
Gamma and the Cauchy distribution respectively.
Solution 7 (10 marks) We obtain the following
"n
−1
i=1 ln(X )+ n
1
P((X1 · X2 · · · · · Xn )n 2
en 2 ∈ [a, b]) = P( √ i ∈ [ln(a), ln(b)])
n

Knowing that for uniformly distributed Xi , the random variable − ln(Xi ) follows an exponential
distribution with parameter λ = 1, we use E(ln(Xi )) = −1 and Var(ln(Xi )) = 1 and the the
independence of (Xi )∞
i=1 to conclude with the Central Limit Theorem that

−1 1
lim P((X1 · X2 · · · · · Xn )n 2
en 2 ∈ [a, b]) = N (0, 1)([ln(a), ln(b)]).
n→∞

16

This study source was downloaded by 100000816115074 from CourseHero.com on 05-13-2023 08:44:53 GMT -05:00

https://www.coursehero.com/file/83506358/Assignments-2015pdf/
 Assignment Part II
Statistics for quantitative finance
Autumn 2015

• This Assignment Part II is individual, containing two questions.

• The submission date on Monday 15/06/2015 before the exam at 6:00pm.

• All problems shall be addressed using R. The code and plots shall be submitted on paper

• The weight of the Assignment Part II is 20%.

Problem 1 (15 marks)

a) Generate a sample of size 1000 consisting of realizations of independent N (4, 1)-distributed

random variables.

b) Write a code for a function which gives the density for this sample depending on the parame-
tres (µ, σ) ∈ R × R+ standing for the mean and standard deviation of the normal density.

c) Apply a numerical maximization procedure to determine a maximum likelihood estimator

for the unknown parameters.

d) Produce a plot which shows where the maximum occurs

Problem 2 (15 marks)

a) Generate a sample X = (Xi )ni=1 size n = 1000 consisting of realizations of independent

random variables following a distribution with density function
4
f (x) = x(2 − x2 )1[0,1] (x), x ∈ R.
3

b) Determine the sample mean and the sample variance for the above realization of X .
1 "j
c) Plot the function j '→ j i=1 Xi for j = 1, . . . , 1000 and interpret the result using Strong
Law of Large Numbers and Central Limit Theorem.

This study source was downloaded by 100000816115074 from CourseHero.com on 05-13-2023 08:44:53 GMT -05:00

https://www.coursehero.com/file/83506358/Assignments-2015pdf/
Powered by TCPDF (www.tcpdf.org)