Review Materials for Final
Coverage: Required Chapters 1.1-1.3, 2.1-2.4, 3.1-3.4, 4.1-4.4, 5.1-5.2, 6.1-6.3
Optional Chapters for extra credit 10.1, 11.1
Suggested Extra Problems from your textbook:
Chapter 2.1-2.4
2.1, 2.3, 2.7, 2.21, 2.23, 2.25, 2.26, 2.33, 2.43, 2.47, 2.68
Chapter 10.1 (for extra credit only)
10.6, 10.11, 10.33, 10.37
Chapter 11 (for extra credit only)
11.1 (b-e), 11.9 (a, e)
Any previously assigned extra problems from the previous review sheets
Final Details
The final is worth 120 points spread over 7 questions. The eighth question is 10 points of extra credit. The breakdown is approximate since I have not written the final yet:
Chapter 1.1-1.3 10 points
Chapter 2.1-2.4 20 points
Chapter 3.1-3.4 10 points
Chapter 4.1-4.2 10 points
Chapter 4.3-4.4 20 points
Chapter 5.1-5.2 20 points
Chapter 6.1 10 points
Chapter 6.2-6.3 20 points
Extra Credit for Chapter 10.1/11: 10 points
Bring a calculator, writing instruments, identification, and a formula sheet (both sides allowed). You are not allowed to eat during the final, you may bring something to drink (non-alcoholic) however.
I am not allowed to reveal grades via e-mail or phone. If you want to know yours as soon as possible, leave a grade card with me. Otherwise, the grades will be posted in a timely manner.
What follows are 11 problems, the final is not this long, it’s just extra practice. Good luck.
1. A
poll on women's issues interviewed 1,025 women and 472 men randomly selected
from the United States. The poll found that 47% of the women said they do not
get enough time for themselves.
(a) Construct
an exact 90% confidence interval for the percentage of women who say they do
not get enough time for themselves.
(b) Your
friend is taking Statistic 11 next year and unfortunately they are enrolled at
8am with Professor Lew and they ask you for help all of the time. Explain to
your friend why we can't just say that 47% of all adult women in the U.S. do
not get enough time for themselves.
2. The following comes from a recent article
in The Wall Street Journal :
Generation X-ers Aren't Relying On the
Survival of Social Security
BY JOHN SIMONS
According to the most recent Wall Street
Journal/NBC News poll, only 39% ofX-ers believe that Social Security will still
be able to provide benefits when they retire. That compares to recent surveys
of all Americans which show that 45% think so.
Assume that 45% figure is a stable, long-run,
historical fact about American beliefs about Social Security benefits. Also
assume that the survey of Generation X-ers had 121 respondents.
A. Test
the hypothesis that the belief that Social Security will still be able to
provide retirement benefits has decreased over time. State the null and the
alternative, perform a test, and state a p-value. Please use a 5% level of
significance as your decision rule. On the basis of your test results, do you
think that Generation X-ers are like other Americans in their beliefs about
social security or are they different?
B. Suppose
in a few years the Wall Street Journal decided to replicate this study
(i.e.draw a new sample) on Generation Y-ers (that's you all, I think...). Let's
assume that the 39% figure is now the stable, long run fact about belief in
Social Security benefits by Americans.
What is the chance that a sample of 64 will
have at least 30% of the surveyed Generation Y-ers believing in Social
Security?
3.
A simple random sample of 100
stocks was drawn from the entire market. The average return was 13%, and the SD
was 6%; furthermore, the distribution of percentage returns in the sample was
close to normally distributed.
Based on these data, is
possible to construct a 95% confidence interval for the percentage of stocks in
the market as a whole that had percentage returns greater than 20%?
Please answer yes or no. If you answer yes, please construct a confidence interval and please explain why this is possible to do. If you answer no, please explain why it is inappropriate to construct a confidence interval given this information.
4. 1996 was a particularly
good year for the stock market.
For 1996 as a whole, the mean return on all common stocks on the NYSE
(New York Stock Exchange) was m = 16.4%.
The standard deviation was about s = 36%.
Assume the distribution of annual returns is roughly normal.
(a) Suppose you selected 9 stocks at random from the NYSE stocks in 1996. What is the expected value (mean) and standard deviation of the returns of randomly chosen portfolios of 9 stocks?
(b)
What percentage of such
portfolios of 9 stocks will lose money (i.e. have returns of zero or less)?
5. Here are two statistics on all persons who consider
themselves investment bankers in 1997:
$820,000 dollars per year
$141,000 dollars per year
Which one of these numbers is
the mean salary from investment banking and which one is the median salary from
investment banking in 1997? Assume
the samples were of good quality.
The mean is
_______________________________
The median is______________________________
Explain your choice in the
space below. Be brief. This is not a long answer.
6. You got a job at an
insurance company. The company
advertises that it processes 90% of its claims on time, that is, within 5
working days of initial receipt. The average processing time of all claims is
2.7 days with a standard deviation of 0.6 days. You have been asked to select
and audit a simple random sample (SRS) of 42 of the tens of thousands of claims
received in the past month. The audit reveals that 37 of the 42 claims were
processed within 5 working days with an average of 2.9 days and a standard
deviation of 1.1 days.
(a) What are the mean and standard deviation for the
number of claims processed on time?
(b) An angry client thinks the processing time of 2.7 days
is a lie and in reality, the company is slower than it advertises. Please test the hypothesis that the
true processing time is 2.7 days against the client's alternative. Is there sufficient evidence to reject
the null at alpha=.05?
7.Here is some Stata output:
.
regress totgross openday
Source | SS df MS
Number of obs =
907
---------+------------------------------
F( 1, 905) = 278.78
Model | 4.9206e+17 1
4.9206e+17
Prob > F = 0.0000
Residual
| 1.5974e+18 905 1.7651e+15
R-squared
= 0.2355
---------+------------------------------
Adj R-squared = 0.2347
Total | 2.0895e+18
906 2.3062e+15
Root MSE
= 4.2e+07
------------------------------------------------------------------------------
totgross
|
Coef. Std. Err. t P>|t| [95%
Conf. Interval]
---------+--------------------------------------------------------------------
openday | 23219.24
1390.652
16.697 0.000
20489.96
25948.51
_cons | 714478.1 2426230 0.294 0.768 -4047214 5476170
------------------------------------------------------------------------------
totgross is Total gross receipts for a movie
openday was the number of
screens the movie was being shown on its first day of viewing
a.
Write out the regression
equation estimated by Stata.
Identify the slope and intercept clearly.
b.
What is the
interpretation of the coefficient for the variable openday? Please state it in plain English and
include how it relates to totgross.
c.
What is the correlation
for totgross and openday? Please
interpret its value in terms of both strength and direction.
d.
A movie titled "How
the Grinch Stole Christmas" opened on 3,127 screens across the United
States on November 17, 2000. From
this information, can you give me an estimate of its total gross receipts? Answer yes or no and then give me a
numerical solution.
e.
Suppose I told you that
a movie titled "Miss Congeniality" had total gross receipts of
$103,271,534. From this
information, can you tell me how many screens it was being shown on its opening
day of December 22, 2000? Answer
yes or no. If you answer yes,
please give a numerical solution and show your work. If you answer no, please explain why this is not possible.
8. High Bias and High Variance are both considered undesirable features of certain sample statistics (such as a sample mean for example). You are working with a team on a marketing study, a sample of size 100 is drawn. One of the variables you are interested in is the average time spent on the internet on any day. You plan to construct confidence intervals and perform some unspecified hypothesis tests. Studies always have problems, and today you have your choice: High Bias or High Variance. Which one would you rather deal with and why?
9. You work for an investment bank and you are evaluating
two companies A and B. Both of
them are in biotechnology and each company has invested a significant amount of
money in a number of projects.
Here is some information on the two companies:
Company A currently has 10 projects under development,
each project cost the company $100,000.
Suppose when completed, each project has the following distribution: a
50% chance that the project will be worth $500,000, a 20% chance that it will
be worth $100,000 otherwise, it will be worth nothing.
Company B currently has 20 projects under development,
each project cost the company $200,000.
Suppose each project when completed has the following distribution: a
60% chance that the project will be worth $400,000 otherwise, it will be worth
nothing.
Assume the projects are independent and when completed
and with the price of the initial investment factored in, the total of the
projects will represent the total net worth of each company.
(a)
What are the expected net
worths for Company A and Company B?
(b) What are the Standard Deviations for Company A and for
Company B?
( c ) Your bank is trying to determine whether a
merger of the two companies is worth the time and expense. They have called you in to make the
call, here is the information you need: the bank thinks the merger should occur
if there is a greater than 25% chance that the combined net worth of the
companies exceeds $5,000,000.
Answer “merge” or
“do not merge” and show the work that lead you to your
decision.
10. Over his career, a basketball player has made
1,210 free throws and missed 214.
The basketball player is about to shoot the ball again, what is his
estimated probability of making a free throw (i.e. getting the ball into the
basket)?
11. A manufacturer of “zip” floppies advertises that 95 out of 100 floppies will have no problems. Suppose you buy a pack of 16 from a discounter and find that one fails to work. If the manufacturer's claim is true, what is the probability that one or more floppies fail in a pack of 16?