Ans8 1

Download as pdf or txt
Download as pdf or txt
You are on page 1of 2

STATISTICS 241.

3
SOLUTIONS TO ASSIGNMENT #8

1. The random variable X takes values 10, 15 and 20 with probabilities 1/4, 1/2 and
1/4 respectively. Hence E[X] = ip(i) = 15 and E[X 2 ] = (100)(1/4) + (225)(1/2) +
P
(400)(1/4) = 950/4. Var[X] = 12.5.
2. Let X denote the number of defective items. Then X ∼ HG(n1 = 4, n2 = 16, r = 3).
Hence E[X] = (r · n1 )/(n1 + n2 ) = 3/5.
3. E[X] = q/p and Var[X] = q/p2 .
4. Let A be the event that the first six births are boys, and B be the event that the six
out of the ten births are boys. Then

P (A ∩ B) P (A) (1/2)10 1
P (A|B) = = = 10
 = 10
.
P (B) P (B) 6 (1/2) 0
1
6

5. Let X denote the number of persons who decide correctly in a 3-person jury. Then
X ∼ Bin(3, p)

P (3-person jury decides correctly) = P (X = 2) + P (X = 3)


   
3 2 3 3
= p q+ p
2 3
= 3p2 − 2p3 .

P (single jury decides correctly) = p.

Therefore a 3-person jury is preferable to a single jury if

3p2 − 2p3 > p ⇒ 3p − 2p2 > 1 ⇒ (p − 1)(1 − 2p) > 0 ⇒ p > 1/2,

i.e., a 3-person jury is preferable when p > 1/2.


6. Note that the probability that they reach a decision on a certain round is 32 p2 q = 3pq.


i. Let X denote the number of rounds needed to reach a decision, then X ∼


Geo(3pq). Hence P (X > n) = 1 − (1 − 3pq)n−1 .
ii. We need to find the minimum n such that P (X ≤ n) ≥ 0.95. Now 1−P (X > n) ≥ 0.95
⇒ P (X > n) ≤ 0.05 ⇒ (1 − 3pq)n ≤ 0.05. Given that p = 1/2, we have

1 1 n 1
(1 − 3 · · ) ≤ 0.05 ⇒ ( )n ≤ 0.05
2 2 4
1
⇒ n log ≤ log(0.05) ⇒ n ≥ 2.16.
4

Therefore the minimum number of tosses required to reach a decision is 3.


7(i).
P ( series ends in seven games) = P (Clay wins the series) + P (Matt wins the series)
   
6 5 2 6
= (0.58) (0.42) + (0.42)5 (0.58)2 = 0.24.
4 4
7(ii).
P ( Clay wins)
P (Clay wins|series ends in seven games) =
P (series ends in seven games)
6
 5 2
4 (0.58) (0.42)
= = 0.71.
0.24
8. Note that E[(X − a)2 ] = E[X 2 + a2 − 2aX] = E[X 2 ] + a2 − 2aE[X]. Let f (a) =
E[(X − a)2 ]. To find the minimum of f (a), set f 0 (a) = 0 to obtain the critical point
a = E[X]. Now f 00 (a) = 2 > 0 implies that when a = E[X], f (a) is minimum. Hence
the result.
9. Given Y = λX1 +(1−λ)X2 , Var[Y ] = λ2 σ12 +(1−λ)2 σ22 . To find the minimum variance
of Y , think of Var[Y ] as a function of λ, and then find the critical point. Make sure
that the function attains minimum at this critical point. Using this minimal point
calculate the minimum variance. The minimum variance is
2 2
σ22 σ12
 
2 2
σ1 + σ2 .
σ12 + σ22 σ12 + σ22
10.
X −µ 1
E[ ] = (E[X] − µ) = 0.
σ σ
X −µ 2 1
E[( ) ] = 2 E[X 2 ] + µ2 − 2µE[X] = 1
σ σ
X −µ
Var[ ] = 1.
σ
11. Var[X + 4] = 16, E[X] = 6 and Var[X] = 16.
P
12. P
Y is a sum of independent random variables. Therefore, E[Y ] = µi and Var[Y ] =
σi2 . Note that EZi = 0 and Var[Zi ] = 1. Hence E[V ] = 0 and Var[V ] = n.
13.
i. φ(t) = (1/3)(et + e2t + e3t .
ii. φ(t) = e5t .
iii. φ(t) = (4/10)et + (3/10)e2t + (2/10)e3t + (1/10)e4t .
14.
i. This is a m.g.f. corresponding to binomial random variable with parameters
n = 12 and p = 1/4. Hence E[X] = 3 and Var[X] = 9/4. You can calculate the
required probabilities.
ii. It is the m.g.f. of a Poisson random variable with λ = 4.6.
15. E[X] = 2, Var[X] = 4/5 and the probability mass function is given as: p(1) = 2/5,
p(2) = 1/5 and p(3) = 2/5.

You might also like