Scribd
Upload
24 views2 pages

2016 Spring 600 Midtermtwosols

This document contains solutions to problems from a midterm exam on probability and statistics. It includes the following: 1) Solutions to coin tossing problems involving computing expectations, variances, and normal approximations. 2) Solutions involving die rolls and computing moment generating functions. 3) Solutions involving uniform random variables and finding minimum values and distributions. 4) Solutions involving a Poisson process for customer arrivals at a pastry shop. 5) Solutions involving exponential distributions and finding minimum values. 6) A solution involving Cauchy distributions and computing a probability.

Uploaded by

Ed Z
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
24 views2 pages

2016 Spring 600 Midtermtwosols

This document contains solutions to problems from a midterm exam on probability and statistics. It includes the following: 1) Solutions to coin tossing problems involving computing expectations, variances, and normal approximations. 2) Solutions involving die rolls and computing moment generating functions. 3) Solutions involving uniform random variables and finding minimum values and distributions. 4) Solutions involving a Poisson process for customer arrivals at a pastry shop. 5) Solutions involving exponential distributions and finding minimum values. 6) A solution involving Cauchy distributions and computing a probability.

Uploaded by

Ed Z
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/ 2

18.

600 Midterm 2, Spring 2016: Solutions

1. (20 points) Consider a sequence of independent tosses of a coin that is biased so that it
comes up heads with probability 2/3 and tails withPprobability 1/3. Let Xi be 1 if the ith
toss comes up heads and 0 otherwise. Write Sn = ni=1 Xi .

(a) Compute E[X1 ] and Var[X1 ]. Answer: 2/3 and 2/9

(b) Compute E[Sn ] and Var[Sn ] as functions of n. Answer: (2/3)n and (2/9)n

(c) Compute the covariance of S5 and S10 . Answer:


5
X 10
X  X5 X
10
Cov Xi , Xj = Cov(Xi , Xj ).
i=1 j=1 i=1 j=1

P5
Terms with i 6= j are zero, so this is i=1 Cov(Xi , Xi ) = 5Var(Xi ) = 10/9.

(d) Using a normal approximation,


Ra estimate the probability that S300 220. You may
2
use the function (a) = 2 ex /2 dx in your answer. Answer: (a) where
1
p p p
a = (220 200)/ 300 2/9 = 20 9/600 = 3600/600 = 6.

2. (20 points) Suppose that X and Y are the outcomes of independent fair die rolls. So
each takes a value in {1, 2, 3, 4, 5, 6}, with all values being equally likely. Write Z = X + Y .

(a) Compute the moment generating function for X. Answer:

et + e2t + e3t + e4t + e5t + e6t


MX (t) = E[etX ] = .
6

(b) Compute the moment generating function for Z. Answer:


 et + e2t + e3t + e4t + e5t + e6t 2
MZ (t) = [MX (t)]2 = .
6

(c) Compute E[Y |Z]. (That is, express the random variable E[Y |Z] as a function of the
random variable Z.) Answer: E[Y |Z] + E[X|Z] = E[Z|Z] = Z and by symmetry
E[Y |Z] = E[X|Z] so E[Y |Z] = Z/2. Alternative answer: check by hand that
given Z = k {2, 3, . . . , 12}, conditional law of Y is uniform on a set of consecutive
integers centered at k/2.

3. (20 points) Let X be a uniformly random variable on [0, 5]. Let Y be an independent
uniformly random variable on [0, 10]. Write Z = min{X, Y }.

(a) Compute the joint density function f (x, y) for X and Y . Answer: f (x, y) = 1/50 if
(x, y) [0, 5] [0, 10], and 0 otherwise.

(b) Compute P (Z > 0) and P (Z > 3) and P (Z > 5). Answer: P (Z > 0) = 1 and
P (Z > 3) = P (X > 3)P (Y > 3) = (2/5)(7/10) = 7/25 and P (Z > 5) = 0.

1
(c) Compute the cumulative distribution function FZ (a). Answer: For a [0, 5], have
(5a)(10a)
1 FZ (a) = P (X > a)P (Y > a) = ( 5a 10a
5 )( 10 ) = 50 . So

0
a<0
(5a)(10a)
FZ (a) = 1 50 a [0, 5] .

1 a>5

4. (20 points) Alices Pastry Shop is open from 7:00 a.m. until 10:00 p.m. Throughout
those 900 minutes, Alice has an extremely steady business: customers show up according
to a Poisson point process with parameter = 1, where time is measured in minutes.
(That is, the expected number of customers per minute is one.) Let N be the total
number of customers that arrive during the day.

(a) Compute the probability that there are exactly 3 customers during the first three
minutes. Answer: e(t) (t)k /k! = e3 33 /6 = (9/2)e3 .

(b) Write a probability density function for the time it takes from the store opening
until the arrival of the second customer. (Imagine that customers keep arriving after
closing, so that with probability one a second customer comes eventually. In other
words, dont worry about the 900 minute upper bound for this part of the problem.)
Answer: this is with parameters = 1 and = 2. Density function is
f (x) = xex /(2) = xex for x 0 (and zero if x < 0).

(c) Compute E[N ] and Var[N ]. Answer: E[N ] = = 900 and Var[N ] = = 900.

(d) Compute the probability that the entire day goes by without a single customer.
Answer: e900 . (Probability -exponential random variable exceeds T is eT .)

5. (10 points) Suppose that X1 , X2 , . . . , Xn are independent exponential random variables


with parameter = 1.

(a) Write Y = min{X1 , X2 , . . . , Xn }. Compute the density function fY . Answer: This


is exponential with rate = n so fY (y) = ey = neny .

(b) Compute E[Y k ] as a function of n and k. You may assume that n and k are positive
integers. RAnswer: Recall that if X is exponential with parameter 1 we have

E[X k ] = 0 xk ex dx = k!. (This is one of the definitions of the factorial.) Note
that Y has same law as X/n, so E[Y k ] = E[(X/n)k ] = E[X k ]/nk = k!/nk .

6. (10 points) Suppose that X1 , X2 , X3 , X4 , . . . , Xn are independent random variables,


1
each of which has a probability density function given by f (x) = (1+x 2 ) . Compute the

probability that X1 + XP 2 + . . . + Xn n. Answer: Each Xi is Cauchy and we seek the


probability that Z = n1 ni=1 Xi 1. Since Z is itself Cauchy, spinning flashlight story
gives probability P (Z) > 1 = (/4)/ = 1/4. (The angle between segment [(0, 1), (1, 0)]
and horizontal line y = 1 is /4.)

You might also like