Homework 4 (Chapter 4)

Homework 4 (Chapter 4)
1. Given events A and B that may overlap, derive the probability in terms of sets using the
axioms of probability for:
a) neither A nor B
P( (A∪B)’ ) = 1 – P(A∪B) = 1 – ( P(A) + P(B) – P(A∩B) )
b) either A or B, not both (exclusive OR)
P( A∪B – A∩B ) = P(A) + P(B) – 2*P(A∩B)
2. Three bins contain some good and defective components as follows:
Bin Good Bad

1 8 2
2 3 1
3 2 2
Over a long period of time, 20% of components are drawn from bin 1, 30% from bin 2, and
50% from bin 3.
(a) Draw a probability tree.

(b) If a defective part is drawn, what are the probabilities it came out of each bin ? Write
the probability table showing the outcomes, prior, conditional, joint, and posterior
probabilities.
(a) Let G = good and B = bad
0.2 0.3 0.5
Bin 1 2 3
G B G B G B
0.8 0.2 0.75 0.25 0.5 0.5
b)
Prior Conditional Joint Posterior

Hi P(Hi) P(B | Hi) P(B | Hi) P(Hi) P(Hi | B)
Bin 1 0.2 0.2 0.04 0.110
Bin 2 0.3 0.25 0.075 0.205
Bin 3 0.5 0.5 0.25 0.685
where probability of drawing a bad part from any bin is
P(B) = P(Bin1 ∩ B) + P(Bin2 ∩ B) + P(Bin3 ∩ B)

= 0.04 + 0.075 + 0.25
= 0.365
(since P(H ∩ E) = P(E | Hi) P(Hi)) and the posterior probabilities, which are the
probabilities the defective component came from bin 1, 2, or 3 are
P(Bin1 ∩ B) 0.04
P(Bin1 | B) = P(B) = 0.365 = 0.110
P(Bin2 ∩ B) 0.075
P(Bin2 | B) = P(B) = 0.365 = 0.205
P(Bin3 ∩ B) 0.25
P(Bin3 | B) = P(B) = 0.365 = 0.685
3. A screening test is a low-cost way of checking large groups of people for a disease. A more
costly but accurate test shows that 1% of all people have the disease. The screening test
indicates the disease (test positive) in 90% of thoses who have it, and in 20% of those who
do not have the disease (false positive).
(c) Make a table showing the outcomes, prior, conditional, joint, and posterior probabilities.
(d) What percentage of people who test positive don’t have the disease (false +) ?
(e) What percentage of people who test negative do have the disease (false – ) ?
a)
Outcomes Prior Conditional Joint Posterior
Hi P(Hi) P(+ | Hi) P(+ | Hi) P(Hi | +)
P(Hi)
Disease H1 0.01 0.90 0.009 0.043
No H2 0.99 0.20 0.198 0.957

Disease
P(+) = ∑ P(+ | Hi) P(Hi) = 0.009 + 0.198 = 0.207

i
is the probability that a person tests positive.
P(+ | H1) P(H1) 0.009

P(H1 | +) = P(+) = 0.207 = 0.043
P(+ | H2) P(H2) 0.198

P(H2 | +) = P(+) = 0.207 = 0.957
b)
Since P(H1 | +) = 0.043, this means that 4.3% of the people who test positive have
the disease. Since P(H2 | +) = 0.957, this means that 95.7% of the people who test
positive do not have the disease (false-positive).
c)
Redo the table with the conditional probabilities of test negative (i.e. 1 − 0.9 = 0.1
and 1 − 0.2 = 0.8).
Outcomes Prior Conditional Joint Posterior

Hi P(Hi) P(− | Hi) P(− | Hi) P(Hi | −)
P(Hi)
Disease H1 0.01 0.10 0.001 0.0013
No H2 0.99 0.80 0.792 0.9987
Disease
P(−) = ∑ P(− | Hi) P(Hi) = 0.001 + 0.792 = 0.793

i
is the probability that a person tests negative.
P(− | H1) P(H1) 0.001

P(H1 | −) = P(E) = 0.793 = 0.13%
is the probability of a person testing negative having the disease (false-negative)
P(H2 | −) = 99.87%
is the probability of a person testing negative and not having the disease.
4. A disk drive may malfunction with either fault F1 or F2, but not both. The possible
symptoms are the following:
A = { bad writes, bad reads } B = { bad reads }
and F2 is three times as likely to occur as F1. For this type of drive
P(A | F1) = 0.4 P(B | F1) = 0.6

P(A | F2) = 0.2 P(B | F2) = 0.8
Find: P(F1), P(F2), P(A), P(B), P(F1 | A), P(F2 | A), P(F1 | B), P(F2 | B)
By Bayes' Theorem
P(A | F1) · P(F1) P(A | F1) P(F1)
P(F1 | A) = P(A | F ) P(F ) + P (A | F ) P(F ) = P(A)
1 1 2 2
It is given that
P(F2) = 3 P(F1)
and since the total probability is 1
P(F2) + P(F1) = 1
then
3 P(F1) + P(F1) = 1
and so
P(F1) = 0.25 P(F2) = 0.75
Also
P(A) = P(A ∩ F1)+ P(A ∩ F2) = P(A | F1) P(F1) + P(A | F2) P(F2)
= (0.4) (0.25) + (0.2) (0.75) = 0.25
P(B) = P(B ∩ F1)+ P(B ∩ F2) = P(B | F1) P(F1) + P(B | F2) P(F2)
= (0.6) (0.25) + (0.8) (0.75) = 0.75
P(A | F1) P(F1) (0.4) (0.25)

P(F1 | A) = P(A) = 0.25 = 0.4
P(B | F1) P(F1) (0.6) (0.25)

P(F1 | B) = P(B) = 0.75 = 0.2
P(A | F2) P(F2) (0.2) (0.75)

P(F2 | A) = P(A) = 0.25 = 0.6
P(B | F2) P(F2) (0.8) (0.75)

P(F2 | B) = P(B) = 0.75 = 0.8
As a check, note that
P(F1 | A) + P(F2 | A) = 1
P(F1 | B) + P(F2 | B) = 1
Note that these equations are also simpler to use in calculating the complementary probabilities
such as P(F1 | A) and P(F2 | A).

Homework 4 (Chapter 4)

Uploaded by

Copyright:

Available Formats

Homework 4 (Chapter 4)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Homework 4 (Chapter 4)

Uploaded by

Copyright:

Available Formats

Homework 4 (Chapter 4)

P( (A∪B)’ ) = 1 – P(A∪B) = 1 – ( P(A) + P(B) – P(A∩B) )

b) either A or B, not both (exclusive OR)

P( A∪B – A∩B ) = P(A) + P(B) – 2*P(A∩B)

2. Three bins contain some good and defective components as follows:

Bin Good Bad

(a) Draw a probability tree.

0.2 0.3 0.5

Prior Conditional Joint Posterior

where probability of drawing a bad part from any bin is

P(B) = P(Bin1 ∩ B) + P(Bin2 ∩ B) + P(Bin3 ∩ B)

No H2 0.99 0.20 0.198 0.957

P(+) = ∑ P(+ | Hi) P(Hi) = 0.009 + 0.198 = 0.207

is the probability that a person tests positive.

P(+ | H1) P(H1) 0.009

P(+ | H2) P(H2) 0.198

Outcomes Prior Conditional Joint Posterior

P(−) = ∑ P(− | Hi) P(Hi) = 0.001 + 0.792 = 0.793

is the probability that a person tests negative.

P(− | H1) P(H1) 0.001

is the probability of a person testing negative having the disease (false-negative)

A = { bad writes, bad reads } B = { bad reads }

P(A | F1) = 0.4 P(B | F1) = 0.6

and since the total probability is 1

P(A | F1) P(F1) (0.4) (0.25)

P(B | F1) P(F1) (0.6) (0.25)

P(A | F2) P(F2) (0.2) (0.75)

P(B | F2) P(F2) (0.8) (0.75)

As a check, note that

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.