Exam Solutions | Engineering Statistics 2009 | STAT 305, Exams of Statics

Material Type: Exam; Class: ENGINEERING STAT; Subject: STATISTICS; University: Iowa State University; Term: Summer Session I 2009;

Typology: Exams

Pre 2010

Uploaded on 09/02/2009

koofers-user-o4q
koofers-user-o4q 🇺🇸

10 documents

1 / 4

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Name: SOLUTION Statistics 305 Exam 2 part I
June 22nd, 2009
[60 points total]
Problem 1 [12 points total]
Suppose for three independent random variables X, Y and Z, we have that
X ~ N(μ = 10, σμ = 10, σ2 = 4), Y ~ N(μ = 5, σ), Y ~ N(μ = 10, σμ = 5, σ2 = 15) and Z ~ Poisson(μ = 10, σλ = 3).
Now define a two random variables, U = 2X – 2Y + Z + 5 and V = XY2 + ½ Z.
a) Calculate E[U]. [3 pts]
EU = 2(μ = 10, σEX) – 2(μ = 10, σEY) + EZ + 5
= 2(μ = 10, σ10) – 2(μ = 10, σ5) + 3 + 5
= 18
b) Calculate Var(μ = 10, σU). [3 pts]
Var(μ = 10, σU) = 22(μ = 10, σEX) + (μ = 10, σ– 2) 2 (μ = 10, σEY) + EZ
= 4), Y ~ N(μ = 5, σ(μ = 10, σ4), Y ~ N(μ = 5, σ) + 4), Y ~ N(μ = 5, σ(μ = 10, σ15) + 3
= 16 + 60 + 3
= 79
c) Calculate E[V]. [3 pts] **Note this is non-linear so we have to use propagation of error formulas**
EV = (μ = 10, σEX)(μ = 10, σEY)2 + ½ EZ
= 10(μ = 10, σ5)2 + ½ (μ = 10, σ3)
= 250 + 1.5 = 251.5
d) Calculate Var(μ = 10, σV). [3 pts]
2/12
2
Z
V
XY
Y
V
Y
X
V
Var(μ = 10, σV) = (μ = 10, σ(μ = 10, σEY)2)2Var(μ = 10, σX) + [2(μ = 10, σEX)(μ = 10, σEY)]2Var(μ = 10, σY) + (μ = 10, σ½)2 Var(μ = 10, σZ)
= 252(μ = 10, σ4), Y ~ N(μ = 5, σ) + 4), Y ~ N(μ = 5, σ(μ = 10, σ100)(μ = 10, σ25)(μ = 10, σ15) + ¼(μ = 10, σ3)
= 152500.8
pf3
pf4

Partial preview of the text

Download Exam Solutions | Engineering Statistics 2009 | STAT 305 and more Exams Statics in PDF only on Docsity!

Name: SOLUTION Statistics 305 Exam 2 part I

June 22

nd

[60 points total]

Problem 1 [12 points total]

Suppose for three independent random variables X, Y and Z, we have that

X ~ N(μ = 10, σμ = 10, σ

2

= 4), Y ~ N(μ = 5, σ), Y ~ N(μ = 10, σμ = 5, σ

2

= 15) and Z ~ Poisson(μ = 10, σλ = 3).

Now define a two random variables, U = 2X – 2Y + Z + 5 and V = XY

2

+ ½ Z.

a) Calculate E[U]. [3 pts]

EU = 2(μ = 10, σEX) – 2(μ = 10, σEY) + EZ + 5

b) Calculate Var(μ = 10, σU). [3 pts]

Var(μ = 10, σU) = 2

2

(μ = 10, σEX) + (μ = 10, σ– 2)

2

(μ = 10, σEY) + EZ

= 4), Y ~ N(μ = 5, σ(μ = 10, σ4), Y ~ N(μ = 5, σ) + 4), Y ~ N(μ = 5, σ(μ = 10, σ15) + 3

c) Calculate E[V]. [3 pts] Note this is non-linear so we have to use propagation of error formulas

EV = (μ = 10, σEX)(μ = 10, σEY)

2

+ ½ EZ

2

d) Calculate Var(μ = 10, σV). [3 pts]

2

Z

V

XY

Y

V

Y

X

V

Var(μ = 10, σV) = (μ = 10, σ(μ = 10, σEY)

2

2

Var(μ = 10, σX) + [2(μ = 10, σEX)(μ = 10, σEY)]

2

Var(μ = 10, σY) + (μ = 10, σ½)

2

Var(μ = 10, σZ)

2

(μ = 10, σ4), Y ~ N(μ = 5, σ) + 4), Y ~ N(μ = 5, σ(μ = 10, σ100)(μ = 10, σ25)(μ = 10, σ15) + ¼(μ = 10, σ3)

Problem 2 [11 points total]

A metal pressing plant produces two sizes of washer. The “small” washer is such that it can be made

from the “inner” waste left from pressing the “large” washer. So, to maximize efficiency, engineers

designed a press that takes a sheet of metal and presses both sizes simultaneously.

Let X and Y respectively denote the number of defective “small” and “large” washers pressed from a

given sheet of metal. The joint probabilities of X and Y appear in the table below.

a) Fill in the following tables with the marginal probabilities of X and Y. [4), Y ~ N(μ = 5, σ pts]

b) Is the number of defective “small” washers pressed from a given sheet of metal

independent of the number of defective “large” washers pressed from that same sheet

of metal? Provide numerical justification for you answer. [4), Y ~ N(μ = 5, σ pts]

No, because for them to be independent, we need f(X = x, Y = y) = fx(x)fy(y) and for example,

f(X = 1, Y = 1) = 0.21 ≠ f

x

(1)f

y

(1) = 0.4(0.6) = 0.24. This is also true for f(1,2), f(2,1) and

f(2,2).

c) Calculate the conditional probability that there are no defective “small” washers, given that

there are 2 defective “large” washers. [3 pts]

f(X = 0|Y = 2) = f(0,2)/fy(2) = 0.05/0.5 = 0.

X

Y

Y 0 1 2

f(μ = 10, σy) 0.1 0.4 0.

X 0 1 2

f(μ = 10, σx) 0.1 0.6 0.

g) How many sections do I expect to inspect until one fails? [2 pts]

EW = 1/p = 1/0.3233 = 3.

Problem 4), Y ~ N(μ = 5, σ

In a grinding operation, there is an upper specification of 3.150 in. on a dimension of a certain part after grinding.

Below is a Normal Q-Q plot for this dimension of 100 parts after grinding.

a) Can we safely determine that the dimension of this certain part after grinding is normally distributed?

Justify your answer. [3 pts]

Yes, because based on the Normal Q-Q plot we can see a linear trend so we know the data is at

least approximately normally distributed.

b) Suppose that the standard deviation of the dimension for parts of this type ground to any particular mean

dimension μ is σ = 0.002 in. Suppose further that you desire to have no more than 5% of the parts fail to

meet specifications. What is the maximum μ that can be used if this 5% requirement is to be met? [5 pts]

^ 

P Z

P X

c) Now suppose you found the mean to be μ =3 in and we now know σ = 0.9 in. What is the probability that

a sample of 16 observations has a sample mean larger than 3.1? [5 pts]

16

P Z

PX P Z