Math 5010–1, Fall 2006
Solutions to the Final Examination
1. Compute the mass function of the random variable Y whose moment generating function is
1 t 1 −2t 1 5t
MY (t) = e + e + e .
2 6 3
Solution: p(1) = 1/2, p(−2) = 1/6, p(5) = 1/3, and p(x) = 0 otherwise.
2. Consider a random vector (X , Y ). We know that X is exponentially distributed with parameter 1; i.e.,
(
e−x , x ≥ 0,
fX (x) =
0, otherwise.
Let x > 0 be any fixed positive number, and suppose that conditionally on the event {X = x}, Y is
exponentially distributed with parameter 1/x. That is,
1 e−y/x , y ≥ 0,
fY |X (y | x) = x
0, otherwise,
Find the density function of (X 2 , Y 2 ).
Solution: The joint density of (X , Y ) is
−x−(y/x)
e
, if x, y ≥ 0,
fX,Y (x , y) = fX (x)fY |X (y|x) = x
0, otherwise.
Let U = g1 (X , Y ) = X 2 and V = g2 (X , Y ) = Y 2 . The Jacobian of the transformation g is 4xy.
Therefore, −√u−(√v/√u)
e
fX,Y (x , y) √ , if u, v ≥ 0,
fU,V (u , v) = = 4u v
4xy
0, otherwise.
3. You have 4 light bulbs whose lifetimes are independent normal random variables with mean 100 (hours)
and standard deviation 5 (hours). Find the probability that your 4 light bulbs together live for at least
402 hours.
Solution: Let Xi denote the lifetime of the ith light bulb. Then Xi is N (100 , 25) for each i. Therefore,
by independence, X1 + · · · + X4 is N (400 , 100). Thus,
402 − 400
P {X1 + · · · + X4 ≥ 402} = P N (0 , 1) ≥ √ ≈ P {N (0 , 1) ≥ 0.2} ≈ 0.4207.
100
4. Suppose X1 , X2 , X3 and X4 are independent with common mean 1 and common variance 2. Compute
Cov(X1 + X2 , X2 + X3 ).
Solution: The covariance is equal to
Cov(X1 , X2 ) + Cov(X1 , X3 ) + Cov(X2 , X2 ) + Cov(X2 , X3 ) = 0 + 0 + 2 + 0 = 2.
1
5. Let X1 , X2 , . . . , X20 be independent
P20 Poisson random variables with mean one. Use the central limit
theorem to approximate P { i=1 Xi > 15}.
P20
Solution: The expectation of each Xi is 1, and so is the variance. Therefore, E( i=1 Xi ) = 20, and
so is the variance. If we apply the CLT, then
( 20 )
X 15 − 20
P Xi > 15 ≈ P {N (20 , 20) > 15} = P N (0 , 1) > √ ≈ P {N (0 , 1) > −1.1} .
i=1
20
This is 1 − Φ(−1.1) = Φ(1.1) ≈ 0.8643.
