You are on page 1of 5

# 1) Let Z be a standard normal random variable.

## Set X = Z if Z > 2 and X = 0

otherwise. Then E(X), the expectation of X, is equal to: . (Write your answer in the
form x.xxx, rounded up to the third digit.)

The distribution of X has a positive probability mass point at zero. The rest of the
probability is distributed according to the normal density over the interval (2, ). The
point mass does not contribute to the expectation. Consequently, the expectation
results from the integration:
E(X) = 2 x(1/2)e-x2/2dx = (1/2) [-e-x2/2]2 = (1/2) e-22/2 = 0.05399097.
After rounding we get 0.054 as the answer.

2) Let X1, X2, ..., X20 be independent and identically distributed Poisson random
variables with rate = 1. Then the bound that is produced by the Markov inequality to
P(X1+X2+ ... +X20 25), the probability that sum of the random variables is greater
than 25, is equal to . (Write your answer in the form 0.xx, rounded up to the second
digit.)

The Markov inequality states that for a non-negative random variable X (X 0):
P(X x) E(X)/x.

In the current case the sum of the Poisson random variables is non-negative with
expectation E(X1+X2+ ... +X20) = 20 E(X1) = 20. Therefore, P(X1+X2+ ... +X20 25)
20/25 = 0.8. No rounding is required.

3) Many people believe that the daily change of price of a company's stock is a
random variable with mean 0 and variance 2. That is, if Yn represents the price of the
stock on the nth day, then Yn = Yn-1 + Xn, n 1, where X1, X2, ... are independent and
identically distributed random variables with mean 0 and variance 2. Suppose that
the current price of the stock is 100 and 2 = 1. Then the probability that the stock's
price will exceed 107 after the passing of 60 days is approximately equal to: . (Use
the normal approximation. Do not attempt to apply a continuity correction. Write your
answer in the form 0.xx, rounded up to the second digit.)

The total change in the price of the stock in 60 days is the total sum of the day-by-day
changes. The expectation of the total change is zero and the variance is 60. This total
change is added to the current price of 100 to produce X, the future price, that has
expectation E(X) = 100 and variance Var(X) = 60, or Var(X) = 7.745967.

## The random variable is a sum of independent and identically distributed random

variables. Applying the Central Limit Theorem we may approximate the probability of
the event {X > 107} with the aid of the normal distribution:
P(X > 107) P(Z > (107-100)/7.745967) = 1 - P(Z 0.9036961) = 1 - 0.8169217 =
0.1830783,
where Z has the standard normal distribution. We get that the probability of the event
is 0.1830783. The answer after rounding is 0.18.

4) A group of 200 men and 200 women is lined up in random. Let X be the number
of men that has a woman next to them. Then E(X), the expectation of X, is equal to: .
(Use the normal approximation with a continuity correction. Write your answer in the
form xxx.xxx, rounded up to the third digit.)

Let Xi be the indicator of the ith position in the row being occupied by a man and that
either the (i-1), the (i+1) position, or both are being occupied by a woman. The
random variables Xi are equal to 1 if the condition is satisfied and 0 otherwise. The
random variables Xi are defined for all i, 1 i 400. However, the definition is
modified in the case i=1, in which only the position i+1 is examined for the presence
of a woman, and in the case i=400, in which only the position i-1 is examined. The
total number of men that have a woman next to them is i=1400 Xi and the expectation
of the number of such men is E(i=1400 Xi) = i=1400 E(Xi) = i=1400 P(Xi=1).

The probability that the ith position is occupied by a man is 200/400 = 1/2. Given this
event, the probability that the i-1 position is occupied by a woman is 200/399. This is
also true for the conditional probability that the position i+1 is occupied by a women.
The event that is indicated by Xi is a union of the two events that are associated by
positions i-1 and i+1, respectively. The conditional probability of the intersection of the
two events, given that position i is occupied by a man is (200/399)(199/398). It follow
that the conditional probability that Xi = 1 is 2x(200/399) - (200/399)(199/398) =
(200/399)(2-199/398). As a result,
P(Xi = 1) = 0.5x(200/399)(2-199/398) = 0.3759398.
In the case i=1 or i=400 this probability is modified to
P(Xi = 1) = 0.5x(200/399) = 0.2506266.
It follows that the expected number of men with a women next to them is
E(i=1400 Xi) = i=1400 P(Xi=1) = 398x0.3759398 + 2x0.2506266 = 150.1253.
After rounding we get 150.125 as the answer.

5) Let X1, X2, ..., X210 be independent and identically distributed random variables, all
sharing the marginal density fX(x) = c x3(5 - x)2, for 0 < x < 5, and fX(x) = 0 otherwise.
Then P(X1+X2+ ... +X210 615), the probability that sum of the random variables is
less or equal to 615, is approximately equal to . (Use the normal approximation. Do
not attempt to apply a continuity correction. Write your answer in the form 0.xx,
rounded up to the second digit.)

Notice that the random variable with the given density can be represented as X = 5Y,
for Y a random variable with the beta distribution. The parameters of the beta
distribution are a=4 and b=3. The expectation of Y is E(Y) = a/(a+b) = 4/7 and the
variance is Var(Y) = ab/(a+b)2(a+b+1) = 12/728 = 3/98. It follows that E(X) = 5E(Y) =
20/7 and Var(X) = 52Var(Y) = 75/98.

## The expectation of the sum of random variables is

E(X1+X2+ ... +X210) = 210 E(X1) = 210x20/7 = 600
and, due to independence, the variance is
Var(X1+X2+ ... +X210) = 210 Var(X1) = 210x75/98 = 160.7143.
Therefore, the standard deviation of the sum is 160.7143 = 12.67731.

Applying the Central Limit Theorem we get that: P(X1+X2+ ... +X210 615) P(Z
(615 - 600)/12.67731) = P(Z 1.183216) = 0.8816382.
After rounding we get 0.88 as the answer.

## 6) Consider n independent flips of a coin having probability p=0.5 of landing on

heads. Say that a changeover occurs whenever an outcome differs from the one
preceding it. (For example, if n is equal to 5 and the outcome is HHTHT, then there
are 3 changeovers.) Then the expected number of changeovers when n=60 is equal
to: . (Write your answer in the form xx.xxx, rounded up to the third digit.)
Let Xi, i = 2, 3, ..., n, be the indicator of the event that a changeover took place on the
ith flip of the coin. Namely, Xi = 1 if the outcome of the ith flip differes from the
outcome in the (i-1)th flip. The total number of changeovers is i=2n Xi and the
expectation of the number of changeovers is E(i=2n Xi) = i=2n E(Xi) = i=2n P(Xi=1).

The probability of a changeover at the ith flip is 1/2: The probability that two flips of a
coin produce opposite outcomes. Consequently, the expected number of
changeovers is
E(i=2n Xi) = i=2n P(Xi=1) = i=2n (1/2) = (n-1)/2.
The answer, if n=60, is (60-1)/2 = 29.5. No rounding is required.

7) If the joint density of (X,Y) is given by f(x,y) = c/y, for 0 < x < y < 5, and f(x,y) = 0
otherwise. Then Cov(X,Y), the covariance between X and Y, is equal to: . (Write your
answer in the form x.xxx, rounded up to the third digit.)

The value of the constant c is obtained by integration of the density function over the
entire region:
1 = c 050y(1/y)dx dy = c 05 dy = 5c c = 0.2.

The value of the mixed moment E(XY) is obtained by integration of the function
xyf(x,y) over the entire region:
E(XY) = 0.2 050y(xy/y)dx dy = 0.2 05 (y2/2) dy = 0.2 53/6 = 4.166667.

## The value of E(X), the expectation of X, is obtained by integration of the function

xf(x,y) over the entire region:
E(X) = 0.2 050y(x/y)dx dy = 0.2 05 (y/2) dy = 0.2 52/4 = 1.25.
The value of E(Y), the expectation of Y, is obtained by integration of the function
yf(x,y) over the entire region:
E(Y) = 0.2 050y(y/y)dx dy = 0.2 05 y dy = 0.2 52/2 = 2.5.

The covariance is the difference between the mixed moment and the product of the
moments:
Cov(X,Y) = E(XY) - E(X)E(Y) = 4.166667 - 1.25x2.5 = 1.041667.
After rounding we get 1.042 as the answer.

8) Two dies are rolled. Let X be the value of the first die and Y the sum of the two
values. Let M(t,s) = E(etX + sY) be the joint moment generating function of (X,Y). The
value of M(1.7,-1.3) is equal to: . (Write your answer in the form x.xxx, rounded up to
the third digit.)

## The joint moment generating function is

M(t,s) = E(etX + sY) = E(e(t+s)X + s(Y-X)) = E(e(t+s)X)E(es(Y-X)),
since X and Y-X are independent random variables. Y and X are not independent.
Make it independent by X and Y-X.

For the discrete uniform distribution over the integers 1 to 6 we get that that the
moment generating function is equal to:
E(etX) = [et - e7t]/[6(1-et)].
Evaluating this function at the point t+s = 1.7-1.3 = 0.4, and then again at the point s
= -1.3, gives:
M(1.7,-1.3) = M(0.4)M(-1.3) = 5.067125 x 0.06241284 = 0.3162537.
After rounding we get 0.316 as the answer.
9) A total of 50 balls, numbered 1 through 50, are put into 50 urns, also numbered 1
through 50 in such a way that ball i is equally likely to go into any of the urns 1, 2, ...,
i. Then the expected number of the urns that are empty is equal to: . (Write your
answer in the form xx.xxx, rounded up to the third digit.)

Let Xi be the indicator of the ith urn being empty. Namely, Xi = 1 if the urn is empty
and 0 otherwise. The total number of urns that are empty is i=150 Xi and the
expectation of the number of empty urns is E(i=150 Xi) = i=150 E(Xi) = i=150 P(Xi=1).

A ball cannot get into urn i before the ith ball is put in an urn. The probability that the
ith ball is not placed in the ith urn is (1-1/i)=(i-1)/i. The probability that (i+1)th ball is
not placed in the ith urn is i/(i+1), the probability for the (i+2)th ball is (i+1)/(i+2), and
so on. The balls are independent since the distribution for each ball is determined
irrespectively of previous and post placements. Consequently, the probability that the
ith urn is empty at the end of all n attempts is
P(Xi=1) = [(i-1)/i][i/(i+1)][(i+1)/(i+2)]...[(n-1)/n] = (i-1)/n.
It follows that the expected number of empty urns is
E(i=150 Xi) = i=150 P(Xi=1) = i=150(i-1)/50 = (49*50/2)/50 = 49/2 = 24.5
The answer is 24.5. No rounding is required.

10) An urn contains 100 black balls. At each stage, a black ball is removed and a
new ball that is black with probability 0.75 and white with probability 0.25 is put in its
place. The expected number of stages needed until there are no more black balls is: .
(Write your answer in the form xxx.x, rounded up to the first digit.)

Given that i black balls remain in the urn, let Xi be the number of stages that are
required until there are i-1 black balls in the urn, i=100, 99, ..., 1. The distribution of Xi
is geometric with p=0.25.

The total number of stages that are required until there are no black balls in the urn is
i=1100 Xi and the expectation of the number of stages is:
E(i=1100 Xi) = i=1100 E(Xi) = 100/0.25 = 400.