Math331 Fall 2008: HW Solutions for Moment Generating Functions and Distributions, Assignments of Mathematics

The solutions to selected problems from section 11.2 of math331, fall 2008 course. The problems involve finding the moment generating functions and probability distributions of independent geometric, binomial, and poisson random variables.

Typology: Assignments

Pre 2010

Uploaded on 09/02/2009

koofers-user-gd5
koofers-user-gd5 🇺🇸

10 documents

1 / 2

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Math331, Fall 2008
Instructor: David Anderson
Section 11.2 Homework Answers
Homework: pg. 474, #’s 2, 6, 8.
2. From number 9 of Section 11.1 (which you did), the moment generating function of a
geometric RV, Xi, with parameter pis
MXi(t) = pet
1qet,
where q= 1 p, and the above function is only valid if t < ln(q). Therefore, if
X1, X2,...,Xnare all independent geometric RVs with parameter p, we may use Theorem
11.3 to find the moment generating function of X1+X2+···+Xn.
MX1+···+Xn(t) = MX1(t)···MXn(t)
=pet
1qet× · · · × pet
1qet(ntimes )
=pet
1qetn
.
Using Table 3 in the Appendix shows that this is the moment generating function for a
negative binomial RV with parameters nand p. The uniqueness Theorem 11.2 then shows
that X1+···+Xnmust be a negative binomial RV.
6. Let X1be the number of bottles under-filled by machine 1 from 10am to 11am tomorrow,
and X2be the number of bottles under-filled by machine 2 from 10am to 11am tomorrow.
Then, X1, X2are binomial RVs with parameters (100, .15),(80, .15). Therefore, by Theorem
11.4 (which we proved and discussed in class), X1+X2is a binomial RV with parameters
(180, .15). We are interested in finding
P{X1+X2= 27}=180
27 .1527.8518027 =.083.
8. X, Y, Z are independent Poisson RVs with parameters λ1, λ2, λ3. Therefore, by Theorem
11.5 (which we proved and discussed in class), X+Y+Zis Poisson with parameter λ1+λ2+λ3
1
pf2

Partial preview of the text

Download Math331 Fall 2008: HW Solutions for Moment Generating Functions and Distributions and more Assignments Mathematics in PDF only on Docsity!

Math331, Fall 2008 Instructor: David Anderson

Section 11.2 Homework Answers

Homework: pg. 474, #’s 2, 6, 8.

  1. From number 9 of Section 11.1 (which you did), the moment generating function of a geometric RV, Xi, with parameter p is

MXi (t) =

pet 1 − qet^

where q = 1 − p, and the above function is only valid if t < − ln(q). Therefore, if X 1 , X 2 ,... , Xn are all independent geometric RVs with parameter p, we may use Theorem 11.3 to find the moment generating function of X 1 + X 2 + · · · + Xn.

MX 1 +···+Xn (t) = MX 1 (t) · · · MXn (t)

=

pet 1 − qet^

× · · · ×

pet 1 − qet^

(n times )

pet 1 − qet

)n .

Using Table 3 in the Appendix shows that this is the moment generating function for a negative binomial RV with parameters n and p. The uniqueness Theorem 11.2 then shows that X 1 + · · · + Xn must be a negative binomial RV.

  1. Let X 1 be the number of bottles under-filled by machine 1 from 10am to 11am tomorrow, and X 2 be the number of bottles under-filled by machine 2 from 10am to 11am tomorrow. Then, X 1 , X 2 are binomial RVs with parameters (100, .15), (80, .15). Therefore, by Theorem 11.4 (which we proved and discussed in class), X 1 + X 2 is a binomial RV with parameters (180, .15). We are interested in finding

P {X 1 + X 2 = 27} =

. 1527. 85180 −^27 =. 083.

  1. X, Y, Z are independent Poisson RVs with parameters λ 1 , λ 2 , λ 3. Therefore, by Theorem 11.5 (which we proved and discussed in class), X+Y +Z is Poisson with parameter λ 1 +λ 2 +λ 3

and X + Z is Poisson with parameter λ 1 + λ 3. We have

P (Y = y | X + Y + Z = t) =

P (Y = y, X + Y + Z = t) P (X + Y + Z = t)

P (Y = y, X + Z = t − y) P (X + Y + Z = t) = P (Y = y)P (X + Z = t − y)

P (X + Y + Z = t)

=

e−λ^2 λy 2 y!

e−(λ^1 +λ^3 )(λ 1 + λ 3 )t−y (t − y)!

t! e−(λ^1 +λ^2 +λ^3 )(λ 1 + λ 2 + λ 3 )t

=

λy 2 y!

(λ 1 + λ 3 )t−y (t − y)!

t! (λ 1 + λ 2 + λ 3 )t

=

t y

λy 2 (λ 1 + λ 2 + λ 3 )y

(λ 1 + λ 3 )t−y (λ 1 + λ 2 + λ 3 )t−y

=

t y

λ 2 λ 1 + λ 2 + λ 3

)y ( λ 1 + λ 3 λ 1 + λ 2 + λ 3

)t−y .

Therefore, this is the distribution function for a binomial RV with parameters t and p = λ 2 /(λ 1 + λ 2 + λ 3 ).