Distribution Problems
Distribution Problems
SOLUTIONS
The following are a number of practice problems that may be helpful for
completing the homework, and will likely be very useful for studying
for exams.
1. a) Let X and Y be random variables with
E ( X ) = µ X = 20, SD ( X ) = σ X = 7,
E ( Y ) = µ Y = 8, SD ( Y ) = σ Y = 2, Corr ( X, Y ) = ρ = 0.60.
Find E ( 3 X – 5 Y ) and SD ( 3 X – 5 Y ).
E ( 3 X – 5 Y ) = 3 µ X – 5 µ Y = 3 ⋅ 20 – 5 ⋅ 8 = 20.
Var ( 3 X – 5 Y ) = Cov ( 3 X – 5 Y , 3 X – 5 Y )
= 9 σX
2
– 30 σ XY + 25 σ Y
2
= 9 σX
2
– 30 ρ σ X σ Y + 25 σ Y
2
= 9 ⋅ 7 2 – 30 ⋅ 0.6 ⋅ 7 ⋅ 2 + 25 ⋅ 2 2 = 289.
SD ( 3 X – 5 Y ) = 289 = 17.
Var ( 3 X + 4 Y ) = Cov ( 3 X + 4 Y , 3 X + 4 Y )
= 9 σX
2
+ 24 σ XY + 16 σ Y
2
= 9 σX
2
+ 24 ρ σ X σ Y + 16 σ Y
2
1
⇒ ρ = − .
3
2. Suppose that you wish to invest in two stocks which both have a current price of $1.
The values of these two stocks in one month are described by two random variables,
say, X 1 and X 2 . Suppose that the expected values and variances of X 1 and X 2
are µ 1 , µ 2 , σ 12 , and σ 22 , respectively. We also assume that the correlation between
the stocks is given by ρ.
Let c denote your initial investment, which is to be invested in the stocks, and assume
that shares can be bought up to any percentages. Let w denote the percentage of your
investment in stock 1. Finally, let P denote the value of your portfolio (investment)
after a month. Then we have that P = c ( w X 1 + ( 1 – w ) X 2 ), where 0 ≤ w ≤ 1.
a) Find an expression for the expected value of your investment after one month.
b) Find an expression for the variance of your investment after one month.
a) Find the probability that on a given day, the price of a package of Oreo cookies is
higher than the price of a gallon of milk. That is, find P ( Y > X ).
P ( Y > X ) = P ( X – Y < 0 ).
0 − 0.30
P ( X – Y < 0 ) = P Z < = P ( Z < – 1.20 ) = Φ ( – 1.20 ) = 0.1151.
0.25
b) Alex is planning a Milk-and-Oreos party for his imaginary friends. He buys 4 gallons
of milk and 7 packages of Oreo cookies. Find the probability that he paid less than $30.
That is, find P ( 4 X + 7 Y < 30 ).
30 − 30.90
P ( 4 X + 7 Y < 30 ) = P Z < = P ( Z < – 0.68 ) = Φ ( – 0.68 ) = 0.2483.
1.32
4. In post-apocalyptic Neverland, clean water is sold in “2-liter” bottles off the back of a
truck that arrives every day. The arrival time is normally distributed with mean 3:06 pm
and standard deviation 24 minutes. The time it takes for the water to sell out is also
normally distributed with mean 30 minutes and standard deviation 7 minutes. The price
of a “2-liter” bottle fluctuates from day to day according to a Normal distribution with
mean 123 rubles and standard deviation 4 rubles. The amount of water in a “2-liter”
bottle varies from bottle to bottle according to a Normal distribution with mean 65 ounces
and standard deviation 1.6 ounces. Assume that all days, all times, and all bottles are
independent.
a) On April 3, 2020, you buy four “2-liter” bottles of clean water. What is the probability
that it costs you more than 500 rubles?
4 × Price has a Normal distribution with mean 4 × µ Price = 4 × 123 = 492 rubles
and variance 4 2 × ( σ Price ) 2 = 4 2 × 4 2 = 256 ruble 2
( standard deviation 16 rubles ).
500 − 492
P ( 4 × Price > 500 ) = P Z > = P ( Z > 0.50 ) = 0.3085.
16
b) On April 3, 2020, you buy four “2-liter” bottles of clean water. What is the probability
that you get more than 2 gallons (256 ounces) of clean water?
E ( W 1 + W 2 + W 3 + W 4 ) = 65 + 65 + 65 + 65 = 260 ounces.
c) On April 3, 2020, you arrive to the lot where the truck stops at exactly 3:00 pm. What is
the probability that the water is sold out by the time you arrive?
T A has a Normal distribution with mean µ A = 3:06 and standard deviation σ A = 24 min
T S has a Normal distribution with mean µ S = 30 min and standard deviation σ S = 7 min
0 − 36
P ( T A + T S < 3:00 ) = P Z < = P ( Z < – 1.44 ) = 0.0749.
25
d) What is the probability that the truck arrives before 3:00 pm on exactly 3 days in one
week?
0 − 6
P ( T A < 3:00 ) = P Z < = P ( Z < – 0.25 ) = 0.4013.
24
Let X = the number of days in one week when the truck arrives before 3:00 pm.
Then X has a Binomial distribution, n = 7, p = 0.4013.
15 − 14
P( Total > 15 ) = P Z > = P( Z > 1.06 ) = 1 – 0.8554 = 0.1446.
0.9434
6. The distribution of the baggage weights for passengers using a particular airline has
a mean of 20 lbs and a standard deviation of 5 lbs. What is the probability that for
(a random sample of) 100 passengers …
2,100 − 2,000
P( Total < 2,100 ) ≈ P Z < = P( Z < 2.00 ) = 0.9772.
50
OR
2,100
Total < 2,100 ⇒ Average < = 21. Need P( X < 21 ) = ?
100
X −μ
n = 100 – large. Central Limit Theorem: ≈ Z.
σ
n
21 − 20
P( X < 21 ) ≈ P Z < = P( Z < 2.00 ) = 0.9772.
5
100
b) the sample mean weight is within 0.5 lb of the overall mean? That is, what is the
probability that the sample mean weight is between 19.5 and 20.5 lbs?
19.5 − 20 20.5 − 20
P( 19.5 < X < 20.5 ) ≈ P <Z< = P( – 1.00 < Z < 1.00 )
5 5
100 100
a) What proportion of boxes are "underfilled"? That is, what is the probability that the
amount dispensed into a box is less than 16 ounces?
µ = 16.12, σ = 0.20.
16.0 − 16.12
P ( X < 16.0 ) = P Z <
0.20
= P ( Z < − 0.60 )
= 0.2743.
b) Find the probability that exactly 2 out of 9 randomly and independently selected
boxes of cereal contain less than 16 ounces.
Let Y = number of boxes of cereal (out of 9) that contain less than 16 ounces.
Need P( Y = 2 ) = ? P(Y = k )= n Ck ⋅ p k ⋅ (1 − p) n − k
c) Find the probability that the sample mean amount of cereal for a random sample of
9 boxes is less than 16 ounces.
n = 9. Need P( X ≤ 16.0 ) = ?
X −μ
We sample from a normally distributed population. ⇒ = Z.
σ
n
16.0 − 16.12
P( X ≤ 16.0 ) = P Z ≤
0.20
9
= P( Z ≤ − 1.80 )
= 0.0359.
d) Suppose that the machine can be adjusted to change the mean while the standard
deviation remains at 0.20 ounces. What must the mean be so that only 20% of all
the boxes are "underfilled"?
µ = 16.168.
8. The weight of an almond varies with mean 0.051 ounce and standard deviation
0.018 ounce.
a) What is the probability (approximately) that the total weight (of a random sample)
of 60 almonds is less than 3 ounces?
3
Total < 3 ⇒ Average < = 0.05. Need P( X < 0.05 ) = ?
60
X −μ
n = 60 – large. Central Limit Theorem: ≈ Z.
σ
n
0.05 − 0.051
P( X < 0.05 ) ≈ P Z < = P( Z < – 0.43 ) = Φ ( – 0.43 ) = 0.3336.
0.018
60
OR
3 − 3.06
P( Total < 3 ) ≈ P Z < = P( Z < – 0.43 ) = Φ ( – 0.43 ) = 0.3336.
0.1394
b) Determine the sample size (the number of almonds) needed to have the probability
of at least 0.80 that the total weight is greater than 16 ounces.
Total − n ⋅ μ
P( Z > – 0.84 ) = 0.7995 ≈ 0.80. ≈ Z.
n ⋅σ
16 − n ⋅ 0.051
= – 0.84. 0.051 ⋅ n – 0.84 ⋅ 0.018 ⋅ n – 16 = 0.
n ⋅ 0.018
x α−1 e − x θ ,
1
f ( x; α, θ ) = 0 < x < ∞, α > 0, θ > 0.
Γ(α ) θα
Suppose α is known.
~
a) Obtain a method of moments estimator of θ, θ .
~ ~ X
µ = α θ. X = α θ. θ = .
α
n α −1
n n 1 n
f (xi ;θ )
1 ∏x
L( θ ) = ∏ =
Γ( α ) θ α
i exp − ∑ i .
x
θ i =1
i =1 i =1
n 1 n
ln L ( θ ) = – n ln Γ ( α ) – n α ln θ + ( α – 1 ) ∑ ln x i –
θ i∑
xi .
i =1 =1
d nα 1 n 1 n
X
dθ
ln L ( θ ) = –
θ
+ ∑ x i = 0. θ̂ =
nα ∑ i
X =
α
.
θ 2 i =1 i =1
For fun:
X μ αθ
E ( θ̂ ) = E ( ) = = = θ.
α α α
θ̂ is an unbiased estimator for θ.
d) Find Var ( θ̂ ).
σ2
“Hint”: Var ( X ) = .
n
X 1 1 σ2 αθ 2 θ2
Var ( θ̂ ) = Var ( ) = Var ( X ) ⋅ = = .
α α2 α2 n α2n αn
e) Find MSE ( θ̂ ).
θ2
bias ( θ̂ ) = E ( θ̂ ) – θ = 0 and Var ( θ̂ ) = .
αn
θ2 θ2
MSE ( θ̂ ) = E [ ( θ̂ – θ ) = ( bias ( θ̂ ) ) + Var ( θ̂ ) = 0 +
2
⇒ ] 2
= .
αn αn
θ2
Note that MSE ( θ̂ ) = → 0 as n → ∞.
αn
10. Imagine you are selected as a contestant on The Price is Right. The host, Bob Barker,
shows you three boxes with marbles in them. Box #1 contains 2 red, 3 white and 5 blue
marbles. Box #2 contains 5 red, 3 white and 2 blue marbles. Lastly, Box #3 contains
3 red, 4 white and 3 blue marbles. One of Bob’s assistants will pick marbles from one
of the boxes after they are hidden behind a curtain. If you correctly guess which box the
marbles were picked from, you win a brand new car! Which box (or boxes) would you
pick if …
• • • • • • • • • • • • • • •
• • • • • • • • • • • • • • •
Box #1 Box #2 Box #3
a) one white and one blue marble are selected, with replacement?
3 5 5 3
Box #1 P(W B) + P(B W ) = × + × = 0.30 ← largest
10 10 10 10
3 2 2 3
Box #2 P(W B) + P(B W ) = × + × = 0.12
10 10 10 10
4 3 3 4
Box #3 P(W B) + P(B W ) = × + × = 0.24
10 10 10 10
Box #1.
b) one red and one blue marble are selected, without replacement?
2 5 5 2 2
Box #1 P(R B) + P(B R) = × + × = ← largest
10 9 10 9 9
5 2 2 5 2
Box #2 P(R B) + P(B R) = × + × = ← largest
10 9 10 9 9
3 3 3 3 2
Box #3 P(R B) + P(B R) = × + × =
10 9 10 9 10
C1 × 3 C 0 × 5 C1 2
Box #1 2
= ← largest
10 C 2 9
C1 × 3 C 0 × 2 C1 2
Box #2 5
= ← largest
10 C 2 9
3 C1 × 4 C 0 × 3 C1 2
Box #3 =
10 C2 10
c) one red, one white, and one blue marble are selected, without replacement?
2 C1 × 3 C1 × 5 C1
Box #1 = 0.25
10 C 3
5 C1 × 3 C1 × 2 C1
Box #2 = 0.25
10 C 3
3 C1 × 4 C1 × 3 C1
Box #3 = 0.30 ← largest
10 C3
Box #3.