Scribd
Upload
160 views3 pages

Math 3215 Intro. Probability & Statistics Summer '14 Group Quiz 6

This document contains a group quiz with 5 problems related to probability and statistics. The problems cover topics like finding the probability density function of the sample mean using convolutions, approximating probabilities using the central limit theorem, and calculating the expected value and variance of the sample mean. The solutions provided demonstrate how to apply formulas and theorems to solve for probabilities, means, and variances.

Uploaded by

Pei Jing
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
160 views3 pages

Math 3215 Intro. Probability & Statistics Summer '14 Group Quiz 6

This document contains a group quiz with 5 problems related to probability and statistics. The problems cover topics like finding the probability density function of the sample mean using convolutions, approximating probabilities using the central limit theorem, and calculating the expected value and variance of the sample mean. The solutions provided demonstrate how to apply formulas and theorems to solve for probabilities, means, and variances.

Uploaded by

Pei Jing
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Math 3215 Intro.

Probability & Statistics Summer ’14

Group Quiz 6
You may work in groups, ask the instructor for the help, use books or notes or the internet.
Any calculator can be used on this quiz.
1. Let W1 , W2 be independent random variables with the Cauchy distribution defined by having
the Cauchy p.d.f.
1
h(w) = , −∞ < x < ∞.
π(1 + w2 )
This problem will help us to find the p.d.f. of the sample mean Y = (W1 + W2 )/2 using
convolutions. First, by doing a change of variables show that the p.d.f. of Xi = Wi /2, i = 1, 2,
is
2
f (x) = , −∞ < x < ∞.
π(1 + 4x2 )
Let Y1 = X1 + X2 and Y2 = X1 and show that the joint p.d.f. of Y1 , Y2 is
g(y1 , y2 ) = f (y1 − y2 )f (y2 ), −∞ < y1 , y2 < ∞.
Finally show that Z ∞
g1 (y1 ) = f (y1 − y2 )f (y2 ) dy2
−∞
is the p.d.f. of Y1 , and conclude that g1 (y) is the p.d.f. of Y .
Solution: Using the change of variables formula, the p.d.f. of Xi = Wi /2, i = 1, 2, is
2 2
fi (x) = h(2x) · (2x)0 = 2
= ,
π(1 + (2x) ) π(1 + 4x2 )
since w = 2x is the inverse function of x = w/2. Next, the Jacobian corresponding to the
linear change of variables Y1 = X1 + X2 and Y2 = X1 is

∂v1 ∂v1
∂y1 ∂y2 0 1
J = ∂v2 ∂v2 = = −1,
∂y1 ∂y2 1 -1

where v1 (y1 , y2 ) = y2 and v2 (y1 , y2 ) = y1 − y2 . Since X1 and X2 are independent (remember


that W1 , W2 were), from the change of variables formula the joint p.d.f. of Y1 and Y2 is
g(y1 , y2 ) = | − 1|f (y2 , y1 − y2 ) = f1 (y2 )f2 (y1 − y2 ),
as desired. Writing f = fi , for i = 1, 2, to get the marginal p.d.f. for Y1 we integrate away the
Y2 variable, Z ∞ Z ∞
g1 (y1 ) = g(y1 , y2 ) dy2 = f (y2 )f (y1 − y2 ) dy2 .
−∞ −∞

2
2

2. Let X1 , X2 be independent random variables which are exponentially distributed with param-
eters λ1 , λ2 , respectively. Find the p.d.f. of Y = X1 + X2 using the convolution formula
Z b
f ∗ g(x) = f (t)g(t − x) dt.
a

Hint: make the necessary modifications to your argument from Problem #1.
Solution: Let fi (xi ) = λi e−λi xi be the marginal p.d.f. of Xi , i = 1, 2. We have that the p.d.f.
g of Y = X1 + X2 is given by the convolution formula g = f1 ∗ f2 ,

g(y) = f1 ∗ f2 (y)
Z ∞
= f1 (t)f2 (t − y) dt
Z0 ∞
= λ1 e−λ1 t · λ2 e−λ2 (t−y) dt
Z0 ∞
λ1 λ2
= λ1 λ2 e−(λ1 t+λ2 (t−y)) dt. = e−(λ1 t+λ2 (t−y))
0 −λ1 − λ2

3. Approximate P (39.75 ≤ X̄ ≤ 41.25), where X̄ is the mean of a random sample of size 28 from
a distribution with mean µ = 40 and variance σ 2 = 4.
√
Solution: We use the central limit theorem with Z = (X̄ − µ)/(σ/ n),
 
39.75 − 40 41.25 − 40
P (39.75 ≤ X̄ ≤ 41.25) = P √ ≤Z≤ √
(2/ 28) (2/ 28)
= P (−.66 ≤ Z ≤ 3.31)
≈ 1 − (1 − .7454) = .7454 .

2
3

4. A random sample of size n = 18 is taken from a distribution with p.d.f. f (x) = 1 − .5x,
0 ≤ x ≤ 2. Approximate P (.66 ≤ X̄ ≤ .83).
Solution: First note that the mean of the random variable X with p.d.f. f (x) above is found in
R2 R2 R2 2
the usual way, µ = 0 xf (x) dx = 0 x(1 − .5x) dx = 0 (x − .5x2 ) dx = (1/2)x2 − (1/6)x3 |0 =
√
2 − 8/6 = 2/3. Similarly, it is easy to see that σ 2 = .222. Setting Z = (X̄ − µ)/(σ/ n), we
have that Z is approximately standard normally distributed. So

P (.66 ≤ X̄ ≤ .83) = P (−.01 ≤ Z ≤ 1.47)


≈ .9292 − (1 − .5080) = .4372 .

5. Let X equal the weight of individual apple coming out of a particular barrel of apples. Suppose
that E(X) = 24 and Var(X) = 2.2. Let X̄ be the sample mean of a random sample of n = 20
apples. Find E(X̄), Var(X̄), and approximate P (23 ≤ X̄ ≤ 24).
P20 1
Solution: We have E(X̄) = 24 and Var(X̄) = i=1 n2 2.2 = 2.2/20 = .11. Setting Z =
√
(X̄− 24)/( .11), we
 have that Z is approximately standard normal, so P (23 ≤ X̄ ≤ 24) =
P 23−24
√
.11
) ≤ Z ≤ 0 = P (−.3015 ≤ Z ≤ 0) ≈ .5 − (1 − .6179) = .1179 .

You might also like