HW4 Solution

STAT 218 Homework 4 Solution
1. General Addition Rule (5 points)

Solution: See attached paper in the end.
2. True or False (20 points: 5pts/each)
(a) True.
Proof : Denote two complementary events, say A and AC . By definition, AC ∩A =
∅. Therefore, they must be disjoint (i.e. mutually exclusive).
(b) False.
Proof : Given A and AC that are well-defined, i.e. 0 < P (A), P (AC ) < 1 (neither
A nor AC is an empty set). Then, P (A ∩ AC ) = P (∅) = 0 but P (A)P (AC ) > 0.
Therefore, they are not independent. From another aspect, one can show that
P (A | AC ) = 0 6= P (A) which also yields dependence.
(c) False.
Counter example: Can construct two complementary events. Conclusion fol-
lows immediately from part (b).
(d) False.
Counter example: One possible example is the trial of first flipping a fair coin
and then toss a six-sided die. The sample space has 12 outcomes, i.e. S =
{(H, 1), . . . , (H, 6), (T, 1), . . . , (T, 6)}. Let event A = {flip a head} and event B =
{toss an even number}. It can be shown by definition that these two events are
independent.
3. Survey probabilities (20 points: 5 pts/each)
(a) The best way to do this is to make a contingency table.
A: not over 30 AC : over 30 Totals
B: male
B C : female
Total
First fill in what we know.
B: male 70 100
B C : female
Total 160 330
Then, fill in the other cells.
B: male 30 70 100
B C : female 140 90 230
Total 170 160 330
1
Because there are 30 people who are both Male and not over 30, events A and B
are not disjoint.
(b) Solution: A and B are not independent. If they were, the rows would be pro-
portional.
(c) Solution:
P (A or B) = P (A) + P (B) − P (A and B)

170 100 30
= + −
330 330 330
= 240/330
≈ 0.7273
(d) Solution:
P (A or B but not both) = P (A) + P (B) − 2P (A and B)

170 100 30
= + −2×
330 330 330
= 210/330
≈ 0.6363
4. Just guess?!! (20 points: 5 pts/each)
(a) Solution: If there are four choices for each question, based on random guessing
the probability of getting the correct answer is P (correct) = 1/4 = 0.25, and
the probability of getting a wrong answer is P (incorrect) = 1 − 0.25 = 0.75. In
addition, notice that the outcome of each question is independent of the outcome
of another question. So, we can assume independence and apply the Multiplication
Rule for the independent case whenever it is applicable.
P (none of the questions correct)

= P (1st incorrect, and 2nd incorrect, and 3rd incorrect, and 4th incorrect, and 5th incorrect)
= P (incorrect) · P (incorrect) · P (incorrect) · P (incorrect) · P (incorrect)
= 0.755
≈ 0.2373
(b) Solution:
P (all of the questions are correct)

= P (1st correct, and 2nd correct, and 3rd correct, and 4th correct, and 5th correct)
= P (correct) · P (correct) · P (correct) · P (correct) · P (correct)
= 0.255
≈ 0.0009766
2
(c) Solution: Notice that {getting at least one of the questions wrong} is the
complement event of {getting all of the questions correct}. If we denote the
event of interest as A = {getting at least one of the questions wrong}, then AC =
{all questions are correct}. By the Complement Rule, we have
P (A) = 1 − P (AC ) = 1 − 0.0009766 ≈ 0.999
(d) Solution: If we get the first incorrect answer on the fourth question, then we
must have gotten the first three questions all correct.
P (first incorrect answer on the 4th question)
= P (1st correct, and 2nd correct, and 3rd correct, and 4th incorrect)
= P (correct) · P (correct) · P (correct) · P (incorrect)
= 0.25 × 0.25 × 0.25 × 0.75
≈ 0.01172
5. Textbook p.p. 418 #56 (25 points: 10+10+5)

From the question we know that
P (advance) = 0.6, thus P (regular) = 1 − P (advance) = 0.4.
In addition, we have
P (no show | regular) = 0.3, and P ((no show | advance) = 0.05.
(a) Solution:
P (no-shows) = P (no-shows ∩ regular) + P (no-shows ∩ advance)
= P (no-shows | regular)P (regular) + P (no-shows | advance)P (advance)
= 0.3 × 0.4 + 0.05 × 0.6
= 0.15
(b) Solution:
P (advance ∩ no-shows)
P (advance | no-shows) =
P (no-shows)
P (no-shows | advance)P (advance)
=
P (no-shows)
0.05 × 0.6
=
0.15
= 0.20
(c) Solution: Because P (advance | no-shows) 6= P (advance), being a no-show is

NOT independent of the type of ticket a passenger holds. (You may use other
methods to verify independence.)
3
6. Textbook p.p. 418 #61 (10 points)
Solution: This is an application of the Bayes Rule.
Let’s denote the event that the lie detector shows the person is dishonest as +, and
that the lie detector shows the person is honest as −. The probability of interest is
P (no lie | +).
P (no lie ∩ +)
P (no lie | +) =
P (+)
P (+ | no lie)P (no lie)
=
P (+ | no lie)P (no lie) + P (+ | lie)P (lie)
0.15 × 0.95
=
0.15 × 0.95 + 0.65 × (1 − 0.95)
= 0.814
Extra Credit: Bonferroni Inequality (1 extra credit, not required)

Let A and B be two events. Prove the Bonferroni inequality:
P (A ∩ B) ≥ 1 − P (Ac ) − P (B c )
Proof : First note that
1 ≥ P (A ∪ B) = P (A) + P (B) − P (A ∩ B)
= (1 − P (AC )) + (1 − P (B C )) − P (A ∩ B) .
Now cancel 1 from both sides and rearrange, to get
P (A ∩ B) ≥ 1 − P (AC ) − P (B C ) .

HW4 Solution

Uploaded by

Copyright:

Available Formats

HW4 Solution

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

HW4 Solution

Uploaded by

Copyright:

Available Formats

STAT 218 Homework 4 Solution

1. General Addition Rule (5 points)

P (A or B) = P (A) + P (B) − P (A and B)

P (A or B but not both) = P (A) + P (B) − 2P (A and B)

4. Just guess?!! (20 points: 5 pts/each)

P (none of the questions correct)

P (all of the questions are correct)

5. Textbook p.p. 418 #56 (25 points: 10+10+5)

(c) Solution: Because P (advance | no-shows) 6= P (advance), being a no-show is

Extra Credit: Bonferroni Inequality (1 extra credit, not required)

Proof : First note that

Now cancel 1 from both sides and rearrange, to get

You might also like

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