1

Business and Economic Statistics B

2009


REVISION QUESTIONS


SEMESTER II




Please note that these questions are revision questions and may not
be in the same style as the examination questions. There is no
guarantee that all parts of the syllabus are covered.


There are 62 questions here. The questions are in a random order. Brief
answers are included at the end. These are not guaranteed to be correct. If you
disagree with any answer, please check with Mr. Goodhew (Room 479) but
please bring your working to the question.

Please only consult us at our consulting times unless you have made prior
arrangements with one of us.

These questions are based upon a compilation of questions and answers made
by Diane Dancer

2
Q1. In a random sample of 500 voters in NSW, 235 thought that the prime minister was doing a good
job. In a random sample of 800 voters in Victoria, 479 thought the same. Do these data suggest
that the proportion of all voters favouring the prime minister is different in the two states? Use a
significance level of 0.01.

Q2. In a large city, 22% of the households had the afternoon paper delivered to their doors. After the
newspaper conducted an aggressive marketing campaign to increase that figure, a random sample
of 200 households was taken. Of the sample group, 61 households now have the paper delivered.
Can we conclude at the 5% significance level that the campaign was a success? Find the p-value of
the test.

Q3. Use the Lagrange method to find the stationary value of
z = x
2
+ 4y
2
+ 6 subject to 2x - 8y = 20

Q4. Data from the USA was used to estimate a time trend equation and an autoregressive equation to
forecast future values on Gross National Product. In 1989, GNP = 4117.7 and in 1990 GNP =
4155.8. Excel gave the following results.

(i) GNP
t
= 979.90 + 72.39 t and (ii) GNP
t
= 260.93 + 0.93 GNP
t-1

Forecast the value of GNP (using both equations) for 1991 and 1992 when 1950 is t = 1.

Q5. Given the equation c = x - Ax where c and x are 2x1 vectors and A is a 2x2 matrix, find
(a) the c vector if x =
100
50

(
and A =
. .
. .
1 4
3 2

(

(b) the x vector if c =
500
800

(
and A =
. .
. .
1 4
3 2

(


Q6. A manufacturer of ski equipment is in the process of reviewing his accounts receivable. He has
noticed that there appears to be a seasonal pattern. Accounts receivable for the past 4 years are
shown in the table.
Year Q1 Q2 Q3 Q4
1986 106 92 65 121
1987 115 100 73 135
1988 114 105 79 140
1989 121 111 82 163

(a) Calculate the seasonal indices.
(b) Calculate the smoothed series.
(c) Suppose the underlying trend series is given by the equation T
t
= 90.4 + 2.02t.
(d) Forecast the trend values for Q3 1990 and Q1 1991. Assume that 1986 Q1 is the first
period.
(e) Forecast the quarterly accounts receivable for Q3 1990 and Q1 1991.

Q7. A manufacturer of computer chips claims that more than 90% of his products conforms to
specifications. In a random sample of 1000 chips drawn from a large production run, 75 were
defective. Do the data provide sufficient evidence at the 1% level of significance to enable us to
conclude that the manufacturer's claim is true? What is the p-value of this test? BE CAREFUL!!

Q8. Solve the following system of equations using a matrix method.
(a) 3x - 4y = 13
(b) x + 5y = -2
3
Q9. In an effort to explain to customers why their electricity bills have been so high lately and how,
specifically, they could save money by reducing thermostat settings on both space heaters and
water heaters, an electricity company has collected total kilowatt consumption (in kilowatt
hours/100) figures for last years winter months, as well as average thermostat settings on space
heaters (in degrees Celsius) and average thermostat settings on water heaters (in degrees
Celsius).
Model fitting results for: Consumption
Independent
variable
coefficien
t
std.
error
t-
value
sig.leve
l
CONSTANT -51.515 8.618 -5.978 0.001
Space Heaters -0.71 1.635 -0.434 0.6792
Water Heaters 1.687 0.573 2.944 0.0258
Analysis of Variance for the Full Regression
Source SS DF MS F P-value
Regressio
n
304.416 2 152.208 69.503 0.0001
Error 13.1397 6 2.189

(a) Write the estimated equation, including the standard errors and R
2
.
(b) Interpret the value of the coefficient for space heaters.
(c) Test the hypothesis that the model is useful for prediction. Use = 0.05.
(d) It is believed that the slope of the variable, Water heaters, is greater than 1.0. Test this
hypothesis at the 1% significance level.
(e) Find the predicted value for the mean consumption given that space heaters are run at 24
o
C
and the water heater thermostat is set at 52
o
C

Q10. A social scientist takes a sample of 300 workers. The sample individuals are classified into one of
4 groups, referred to as labourer, clerical, professional and other. The numbers classified
in these groups are 91, 108, 76 and 25 respectively. Do these data indicate that the proportions of
labourers, clerical, professional and others are the same. Use = 0.05.

Q11. There are 2 different variations of seasonal dummy variables and a time trend. The first quarter of
data was in Quarter 1 in 1985.
The equations are as follows.
(i) y
t
= 0.419 + 0.025 t - 0.030 D1 - 0.084 D3 - 0.269 D4
(ii) y
t
= 0.025 t + 0.389 D1 + 0.419 D2 + 0.334 D3 + 0.150 D4
Forecast the estimated earnings per share for Q2 1991 and Q4 1992 using Models (i) and (ii).

Q12. A computer firm has an order for 300 Pentium computers and wishes to distribute their
manufacturing between two factories. The total cost of producing outputs q
1
from factory 1 and q
2

from Factory 2 is C = f(q
1
, q
2
) = 2q
1
2
- 6q
1
q
2
+ 7q
2
2
How should the manufacturing be distributed? Show that your solution is a minimum.

Q13. Solve the following system of equations using a matrix method.
i) 3x - 4y = 30
ii) 2y - x = -12

Q14. A life assurance company checks on the proportion of its policies that are cancelled within one year
in 2 regions. Of 200 policies sampled in Region A, 40 were cancelled within the first year, whereas
15 of the 50 policies were cancelled in Region B. Should the company conclude that the
proportion of policies cancelled in Region B is different to that in Region A? Use = 0.05.

4
Q15. The quarterly sales of a department store chain were recorded for the past 4 years.
Year Q1 Q2 Q3 Q4
1986 18 33 25 41
1987 22 20 36 33
1988 27 38 44 52
1989 31 26 29 45

(a) Calculate the centred 4-quarterly moving average.
(b) Calculate the seasonal indices
(c) Calculate the smoothed series.
Suppose the underlying trend series is given by the equation T
t
= 25 + 0.89t.
(d) Forecast the trend values for Q2 1990 and Q3 1991. Assume that 1986 Q1 is the first period.
(e) Forecast the sales for Q2 1990 and Q3 1991.

Q16. Are attitude towards social legislation and occupation independent? Use = 0.05.
Attitude Toward Social Legislation
Occupation Favour Neutral Oppose
Blue-collar 19 16 37
White-collar 15 22 46
Professional 24 11 32

Q17. A manufacturer of electric clothes dryers performs many tests on the machines before they leave
the plant. One such inspection involves measuring the air temperature as it enters the drying drum,
for a given fabric setting. When in control, the process mean is at the engineering specification of
118
o
F, with a standard deviation of .60
o
. If a sample of six dryers is checked every 90 minutes for
their air temperature, what would be the upper and lower control limits of the X-bar chart?
Q18. Quarterly sales data for caterpillar tractors was used to produce the following equations using (i) a
time trend model, (ii) and (iii) 2 different variations of seasonal dummy variables and a time trend.
The first quarter of data was in Quarter 3 in 1986.
The equations are as follows.
y
t
= 1025.02 + 50.61 t
y
t
= 921.40 + 51.34 t + 71.02 D1 + 186.89 D2 + 120.08 D3
y
t
= 51.35 t + 992.41 D1 + 1108.29 D2 + 1041.47 D3 + 921.40 D4
Forecast the estimated sales for tractors for Q3 1992 and Q4 1993 using Models (a), (b) and (c).

Q19. A banks proof operation station utilizes the magnetic ink symbols printed on customer cheques to
process them. If the magnetic sensor is unable to read any of the numbers, the check is rejected and
must be processed manually. To track the process rejection rate, the bank samples 500 checks per
day for several days and records the number that had to be manually processed; see the table.
Determine the process center line and control limits to start a p-chart for the bank, assuming that
the sample size will remain at 500.

Day Number rejected Day Number rejected
1 5 7 4
2 4 8 8
3 10 9 2
4 7 10 5
5 3 11 8
6 3 12 7

5
Q20. A real-estate agent used regression techniques in appraising the value of single-family homes
within a certain community which is divided into 3 neighbourhoods. The following variables were
used:
Y = selling price of a house in dollars, X
1
= total living area (in square feet),
D
2
= 1 if in neighbourhood 1 or 0 otherwise,
D
3
= 1 if in neighbourhood 2 or 0 otherwise,
D
4
= 1 if lot size is larger than the typical house lot or 0 otherwise.

1 2 3 4

= 34185.02 + 20.92 X - 2004.50 D + 4276.73 D + 8613.73 D

(89793.9) (6.89) (629.3) (770.06)
Y


R
2
= 0.8852
Part of the Anova Table
Source df F-ratio Prob value
Regression 4 18.98 0.00012
Error 10
Total 14
(a) Interpret the coefficient of the variable D
4
.
(b) Interpret the coefficient of the variable, living area.
(c) Test whether the model is useful for prediction at = 0.01.
(d) Predict the selling price of a house that is 2000 square feet, is in Neighbourhood 1and has a lot
size larger than the typical house lot.
(e) Test whether there is a positive relationship between the selling price and the total living area at
= 0.01.
(f) Test whether the coefficient of the variable, D2, is negative at = 0.05.

Q21. Officials of City Rail claim that less than 10% of all its trains are late. If a random sample of 70
trains shows that only 60 of them are on schedule, can we conclude that the claim is false? (Use
= 0.10.) Be careful!!

Q22. To determine if a single die is balanced, or fair, the die was rolled 600 times. The observed
frequencies with which each of the 6 sides of the die turned up are recorded in the table. Is there
sufficient evidence to conclude at the 5% level of significance, that the die is not fair?
Face 1 2 3 4 5 6
Observed Frequency 114 92 84 101 107 102

Q23. Find the critical point(s) of the function and show whether the point(s) is a maximum, minimum or
neither. Hence calculate the value of the function at the critical point(s).
f = x
2
+ y
2
+ xy - 9x + 1

Q24. Use the Lagrange method to find the stationary value of
z = 59x + 30y -2x
2
- 3y
2
subject to x + 2y = 22

Q25. Find the critical point(s) of the function and show whether the point(s) is a maximum, minimum or
neither. Hence find the value of the function at the critical point(s).
f = x
2
+ 3y
2
+ 4x - 9y + 3

Q26. Solve the following system of equations using a matrix method.
i) x + 5y = 27
ii) 3x - 2y = -4


6
Q27. Data from the USA was used to estimate a time trend equation and an autoregressive equation to
forecast future values on Consumption. In 1989, Cons = 2656.8 and in 1990 Cons = 2682.2. Excel
gave the following results.

(i) Cons
t
= 500.94 + 49.91 t and (ii) Cons
t
= 137.96 + 0.95 Cons
t-1

Forecast the value of GNP (using both equations) for 1991 and 1992 when 1950 is t = 1.

Q28. A psychology student is interested in the behaviour of prisoners as they pass through the various
stages of their sentences. He collects data on 200 attempted escapes (both successful and
unsuccessful). Each attempt is classified by the quarter of the sentence when it was made; in the
first, second, third or fourth.
Quarter 1 2 3 4
Number of Attempts 59 38 42 61
Do these results appear consistent with the hypothesis that attempted escapes are equally likely to occur in
any quarter? Use = 0.01.

Q29. Last year in an election, a politician received 58% of the votes cast. Several months later, a survey
of 700 people revealed that 54% now supported her. Is this sufficient evidence to conclude that her
popularity has decreased? (Use = 0.05) What is the p-value of this test?

Q30. The tourist industry is subject to enormous seasonal variations. A hotel on the Gold Coast has
recorded its occupancy rates for the past 4 years.

Year Q1 Q2 Q3 Q4
1990 .56 .70 .8 .57
1991 .58 .74 .87 .61
1992 .59 .74 .73 .60
1993 .62 .71 .81 .63

(a) Calculate the seasonal indices.
(b) Calculate the smoothed series.
(c) Suppose the underlying trend series is given by the equation T
t
= .64 +.00525t. Forecast the
trend values for Q4 1994 and Q2 1995. Assume that 1990 Q1 is the first period.
(d) Forecast the hotel occupancy rates for Q4 1994 and Q2 1995.

Q31. Find the critical point(s) of the function and show whether the point(s) is a maximum, minimum or
neither.
f = 4.5m + 5n - 0.5m
2
- n
2
- 0.25mn

Q32. Given the following matrices,
A =
2 0 2
1 3 1
1 1 4

(
(
(
B =
10 3 4
1 0 1
3 0 3

(
(
(
C =
3 1 3
0 7 2

(
D =

(
2 4 4
1 1 3


E =
6 5
4 1
2 3

(
(
(
F =

(
(
(
2 1
6 3
4 2
G = [ ] 4 2 5 H =
0
1
5

(
(
(


calculate (i) A B (ii) BA (iii) C + F (iv) GH (v) HG (vi) A + 3I
(vii) CD (viii) CD (ix) G (x) EF

7
Q33. The following data show the number of business school graduates hired from 2 years ago who are
still with a firm, classified by degree.

Degree Number Hired Number Remaining
Bachelor 205 123
Masters 50 18
Test whether the proportions of bachelor- and masters-degree graduates who remain with the firm for 2
years are equal. Use = 0.01.

Q34. The following question uses the model Y =
0
+
1
X
1
+ . Complete the table. Is the model
useful for prediction at = 0.05?

Source Sums of Squares DF Mean Sums of Squares F ratio
Regression 84.19083 1 A E
Error B C D
Total 210.1564 10

Q35. Find the critical point(s) of the function and show whether the point(s) is a maximum, minimum or
neither.
f = 2lk - l
2
+264k - 10l -2k
2


Q36. The quarterly earnings of a large soft-drink company have been recorded for the past 4 years.
These data are shown in the accompanying table. The seasonal estimates are S
1
= 0.84, S
2
= 1.23,
S
3
= 0.96 and S
4
= 0.97. If the series has values (y
t
) of

Q1 Q2 Q3 Q4
1989 63 66 82 98
1990 67
(a) Is the model additive or multiplicative? State your reasons.
(b) Calculate the seasonally adjusted series for the y values in the table.
(c) Suppose the underlying trend series is given by the equation T
t
= 61.7 + 1.18t.
(d) Forecast the trend values for Q4 1990 and Q2 1991. Assume that 1986 Q2 is the first period.
(e) Forecast the quarterly earnings for Q4 1990 and Q2 1991.

Q37. Use the Lagrange method to find the stationary value of z = x - 3y -xy subject to x + y = 6

Q38. In a TV commercial, the manufacturer of a toothpaste claims that less than 4 out of 5 dentists
recommend its product. In order to test this claim, a consumer group randomly samples 400
dentists and finds that 312 recommended the toothpaste. At the 5% significance level, test to
determine if there is sufficient evidence to support the claim? What is the p-value of this test?

Q39. In a genetics experiment, Mendel crossed contrasting strains of garden peas. The details of the
experiment are below.
Characteristics Probability Outcome
Round and Yellow 9/16 315
Round and Green 3/16 101
Wrinkled and Yellow 3/16 108
Wrinkled and Green 1/16 32
According to Mendels theory, the 4 combinations should have occurred with the probabilities shown in
the table. Do the data support Mendels theory? Test using a chi-square goodness of fit test. Use =
0.05.
8
Q40. A pre-election Gallup poll on voting patterns in the next election resulted in the following table.
Test if the political party preference is the same in each state. Use = 0.05.
State
Political Party NSW VIC QLD WA
Labour 105 120 105 70
Liberal 120 100 130 150
Democrat 25 30 15 30

Q41. Find the critical point(s) of the function and show whether the point(s) is a maximum, minimum or
neither. Hence find the value of the function at the critical point(s)
f = y - y
2
- 3x -6x
2
- 7

Q42. Are the events Income Level and Church Attendance independent? Use = 0.01.
Income Level
Church Attendance Low Middle High
Never 27 48 15
Occasional 29 58 13
Regular 24 74 12

Q43. The Christmas break (December 20 - January 15) is a critical period for the tourist industry in
Tasmania. If a relatively small number of people visit the island in this time, the hotels, restaurants
and other tourist attractions will suffer financial losses. A marketing analyst hired by the Chamber
of Commerce in Hobart to promote more tourism wants to analyse the factors that influence people
to come to Tasmania during the Christmas break. She believes that the crucial factors affecting
hotel vacancy rates are tied to weather conditions in the previous year i.e. the mean daily high
temperature (in degrees Celsius) and the number of rainy days during the Christmas break
Model fitting results for: vacancy
Independent variable coefficient std. error t-value sig.level
CONSTANT 33.314 7.242 4.6 0.0006
Temperature -1.128 0.304 -3.71 0.003
Rainy Days -0.267 0.507 -0.527 0.6085

Analysis of Variance for the Full Regression
Source SS DF MS F P-value
Regressio
n
71.766 2 35.883 7.445 0.0079
Error 57.833 12 4.819

(a) Write the estimated equation, including the standard errors and R
2
.
(b) Interpret the value of the coefficient for rainy days.
(c) Test the hypothesis that the model is useful for prediction. Use = 0.05.
(d) It is believed that the slope of the variable, temperature, is negative. Test this hypothesis at the
1% significance level.
(e) Find the predicted value for the mean vacancy rate given that the temperature is an average of
21
0
C and the average number of rainy days last year was 5.

Q44. Grade inflation occurs as an ever higher proportion of students is given high grades, particularly
HDs. In a random sample of 100 grades this semester, 6 were HDs, 15 were Ds, 20 were Cs, 39
were Ps and 20 were Fs. Is this consistent with the universitys policy that 3% should be HDs,
10% Ds, 25% Cs, and 47% Ps? Use = 0.01.

Q45. Use the Lagrange method to find the stationary value of
9
z = - 2x
2
+ 5y
2
+ 7 subject to 4x - 2y = 9
Q46. The following question uses the model Y =
0
+
1
X
1
+
2
X
2
+ . Complete the table. Test
whether there is any relationship between the x and y variables.

Source Sums of Squares DF Mean Sums of Squares F ratio
Regression 84.19083 2 A E
Error 31.45996 5 B
Total C D

Q47. Solve the following system of equations using a matrix method.
i) 2x + 5y = 8
ii) 3x - 2y = -7
Q48. A toy-manufacturing company has been given a large order for small plastic whistles that will be
given away by a large fast-food hamburger chain with its kids meal. Seven random samples of
four whistles have been taken. The weight of each whistle has been ascertained in grams. The data
are shown here. Use these data to construct an x chart and an R chart. What managerial decisions
should be made on the basis of these findings?

Sample 1 Sample 2 Sample 3 Sample 4 Sample 5 Sample 6 Sample 7
4.1 3.6 4.0 4.6 3.9 5.1 4.6
5.2 4.3 4.8 4.8 3.8 4.7 4.4
3.9 3.9 5.1 4.7 4.6 4.8 4.0
5.0 4.6 5.3 4.7 4.9 4.3 4.5

Q49. Are the events Age Group and Type of car driven independent? Use = 0.05.
Age Group
Type of Car Driven 16 - 21 22 - 30 31 - 45 46+
Sports car 10 10 12 8
Compact 10 6 6 8
Medium 12 13 20 25
Large 8 11 22 19

Q50. Cable TV subscriptions over the past few years have been growing dramatically, although
somewhat erratically. In one state in USA, the number of cable subscribers in thousands was
recorded for the past 4 years. The seasonal estimates are S
1
= 0.94, S
2
= 1.05, S
3
= 1.05 and
S
4
= 0.96.

Q1 Q2 Q3 Q4
1993 200 229 236 219
1994 211
(a) Is the model additive or multiplicative? State your reasons.
(b) Calculate the seasonally adjusted series for the y values in the table.
(c) Suppose the underlying trend series is given by the equation T
t
= 161.2 + 4.79t.
(d) Forecast the trend values for Q1 1996 and Q3 1998. Assume that 1990 Q3 is the first period.
(e) Forecast the quarterly earnings for Q1 1996 and Q3 1998.

Q51. In a test to evaluate the effectiveness of a new drug for combating acne, a drug company found that
120 out of a random sample of 220 acne sufferers improved after being treated with this new
preparation. The leading acne treatment has a 50% improvement rate. Can we conclude from the
data that this new drug is more effective in treating acne? (Use = 0.01)


10

Q52. Test if the style, professional women prefer depends on their occupation. Use = 0.05.
Style
Profession A B C D
Teller 4 10 6 10
Secretary 8 10 15 7
Entrepreneur 7 17 21 15
Socialite 11 13 28 18

Q53. A companys products are classified as perfect, seconds or defective. A random sample of 150
products was classified as follows.
Condition of Product Number
Perfect 104
Second 38
Defective 8
Are these data consistent with the probabilities P(perfect) = 0.8 and P(second) = 0.15 at = 0.05?

Q54. In a random sample of 100 units from an assembly line, 18 were defective. Does this provide
sufficient evidence at the 5% significance level to allow us to conclude that the defective rate
among all units exceeds 10%? What is the p-value?

Q55. The starting salaries and Graduate Management Aptitude Test (GMAT) scores for a sample of 10
MBAs who graduated in 1990 are analyzed and the results produced below.

Regression Analysis - Linear model: Y =
0
+
1
x +
Dependent variable: salary Independent variable: gmat
Independent
variable
coefficient std. error t-
value
Intercept 46425.9 22828.7 A
Salaries -24.1379 41.4274 B

Analysis of Variance
Source SS DF MS F
Regressio
n
6758.62 1 C E
Error 15927 8 D

(a) Complete the two tables.
(b) Is the intercept coefficient significant?
(c) Test the hypothesis that the slope coefficient is greater than -26, using = 0.05.
(d) Is there a relationship between gmat and starting salaries.

Q56. Quarterly sales data for Johnson and Johnson was used to produce the following equations using (i)
a time trend model, (ii) and (iii) 2 different variations of seasonal dummy variables and a time trend
and (iv) an autoregressive model. The first quarter of data was in Quarter 3 in 1986.The equations
are as follows.
y
t
= 1702.25 + 64.15 t
y
t
= 1842.11 + 61.29 t - 76.29 D2 - 164.66 D3 - 221.21 D4
y
t
= 61.29 t + 1842.11 D1 + 1765.81 D2 + 1677.44 D3 + 1620.9 D4
y
t
= 209.88 + 0.94 y
t-1

Forecast the estimated sales for Q3 1990 and Q4 1990 using all models. The sales for Q2 1990
were 2825.
11

Q57. An auctioneer of semi-antique and antique Persian rugs kept records of his weekly auctions in order
to determine the relationships among price, age of carpet, number of people attending the
auction and number of times the winning bidder had previously attended his auctions. He felt
that, with this information, he could plan his auctions better, serve his customers better and make a
higher profit for himself.
Model fitting results for: price
Independent variable coefficient std. error t-value sig.level
CONSTANT 239.532 108.876 2.2 0.0386
Age of carpet 7.37 1.459 5.051 0.0000
Audience size 10.181 2.273 4.479 0.0002
Previous attendance 17.309 9.04 1.914 0.0686

Analysis of Variance for the Full Regression
Source SS DF MS F P-value
Regressio
n
5530975 3 1843658 88.2557 0.0000
Error 459579 22 20890

(a) Write the estimated equation, including the standard errors and R
2
.
(b) Interpret the value of the coefficient for age of the carpets.
(c) Test the hypothesis that the coefficients of the regressors (excluding the constant) are jointly
significant. Use = 0.05.
(d) It is believed that the slope of the variable, Audience size, is less than 15. Test this hypothesis
at the 5% significance level.
(e) Find the predicted value for the mean price given that the age of the carpet is 80 years, the
audience size is 60 and the previous attendance is 5.


Q58. A university mathematics department has put together an exam for students who have had calculus
in high school. Those students who pass the test will be given 3 hours of university credit. For
security purposes, the department wishes to develop and rotate three different versions or forms of
the exam. A prime consideration is to have an equivalent degree of difficulty for all tests. If 120
randomly assigned students now have taken the test versions with the results given here, is it
reasonable ( = .05) to conclude that the objective of equivalent forms has been achieved?

2 . 68 , 8 . 65 , 6 . 67
3 2 1
= = = X X X
20 . 6074 , 40
3 2 1
= = = = SSE n n n
(N.B. SSE = SSW)

Q59. A large manufacturer makes valves. Currently it is producing a particular valve for use in
industrial engines. As a part of a quality control effort, the company engineers randomly sample
seven groups of 40 valves and inspect them to determine whether they are in or out of compliance.
Results are shown here. Use he information to construct a p chart. Comment on the chart.

Sample Size Number not in
Compliance
1 40 1
2 40 0
3 40 1
4 40 3
5 40 2
6 40 5
12
7 40 2

Q60. To predict the asking price of a second hand Holden, data were collected on the cars age, condition
(rated as poor, average or excellent) and mileage and on whether the seller is an individual or a
dealer. The variables were:
y = asking price ($) X
1
= age (in years) X
2
= mileage in thousands
D
3
= 1 if the condition of the car is average and 0 otherwise
D
4
= 1 if the condition of the car is poor and 0 otherwise
S
5
= 1 if the seller is a dealer and 0 if an individual.

. Y S = 17357.4 - 1131.9 X - 33.2 X - 2556.4 D - 3275.3 D

(1280.3) (274.8) (23.6) (915.1) (1112.0) (913.1)
3 4 1 2 5
7756


R
2
= 0.8965
Part of the Anova Table
Source df F-ratio Prob value
Regression 5 43.29 0.0000
Error 25
Total 30
(a) Interpret the coefficient of the variable S
5
.
(b) Interpret the coefficient of the variable, age.
(c) Test whether the model is useful for prediction at = 0.01.
(d) Predict the asking price for a car thats 6 years old with 90,000 (Be careful!!)on the clock, in
excellent condition and sold by a dealer.
(e) Predict the asking price for a car thats 5 years old with 60,000 (Be careful!!)on the clock, in
poor condition and sold by an individual.
(f) Test whether there is a negative relationship between the asking price and the mileage at =
0.01.
(g) Test whether the coefficient of D
4
is negative at = 0.05. Be careful with the conclusion.

Q61. Below is the output for a one-way ANOVA. Analyze the results. Include the number of treatment
levels, the sample sizes, the F value, the overall statistical significance of the test, and the values of
the means.

One way Analysis of Variance
Analysis of Variance

Source Df SS MS F P
Factor 3 1701 567 2.95 0.040
Error 61 11728 192
Total 64 13429


Level N Mean StDev
C1 18 226.73 13.59
C2 15 238.79 9.41
C3 21 232.58 12.16
C4 11 239.82 20.96


13
Very Brief Answers to Revision Questions

Semester II



Q1. Ho: p
1
p
2
= 0 H
1
:
1

2
0 obs
2
= 20.6 Rej H
o
pvalue =5.65E06
Q2. Ho: p = 0.22 H
1
: p > 0.22 obs Z = 2.90 Rej H
o
pvalue = 0.0019
Q3. x = 2 y = -2
Q4. i) 1991 GNP = 4020.28 1992 GNP = 4092.67
ii) 1991 GNP = 4125.82 1992 GNP = 4097.94
Q5. i) c =
|
\

|
.
|
70
10
ii) x =
|
\

|
.
|
1200
1450

Q6. a) 1.0920, 0.9629, 0.6941, 1.2510
c) 128.78, 132.82 d) 89.39, 145.04
Q7. Ho: p = 0.9 H
1
: p > 0.9 obs Z = 2.635 Rej H
o
p value = 0.0041
Q8. x = 3, y = -1
Q9. i) y = -51.515 - 0.71 sp.heat + 1.687 water heat
(8.618) (1.635) (0.573) R
2
= 0.96
iii) Ho: model not useful H
1
: model useful obs F = 69.5 Rej H
o

iv) H
0
:
2
= 1 H
1
:
2
> 1 obs t = 1.199 Dont Rej H
o
v) 19.169
Q10. H
0
: p
1
= p
2
= p
3
= p
4
= 0.25 H
1
: H
0
not true obs X
2
= 51.28 Rej H
o

Q11. i) Q2 1991 y
t
= 1.069 Q4 1992 y
t
= 0.95
ii) Q2 1991 y
t
= 1.069 Q4 1992 y
t
= 0.95
Q12. q
2
= 100 q
1
= 200
Q13. x = 6 y = -3
Q14. Ho: p
2
p
1
= 0 H
1
:
2

1
0 obs
2
= 2.33 Dont Rej H
o
pvalue =0.127
Q15. b) 0.8124, 0.8332, 1.0748, 1.2797
d) 41.02, 45.47 e) 34.18, 48.87
Q16. H
0
: attitude and occupation independent H
1
: not indep. obs X
2
= 6.61 Dont Rej H
o

Q17. LCL = 117.265, UCL = 118.735
Q18. Q3 1992 i) 2290.27 ii) 2324.98 iii) 2325.22
Q4 1993 i) 2543.32 ii) 2461.6 iii) 2461.9
Q19. Centre line 0.011, UCL = 0.025
Q20. c) H
0
:
1
=
2
= .. =
4
= 0 H
1
: H
0
not true obs F =18.98 p value = 0.00012 Rej H
0

d) y = $82 634.25 e) H
0
:
1
= 0 H
1
:
1
> 0 obs t = 3.036 Reject H
0

f) H
0
:
2
= 0 H
1
:
2
> 0 obs t = -3.185 Reject H
0

Q21. H
0
: p = 0.1 H
1
: p > 0.1 obs Z = 1.196 Dont Rej H
o

Q22. H
0
: proportions all equal to 1/6 H
1
: H
0
false obs X
2
= 5.70 Dont Rej H
o

Q23. (6, -3) minimum
Q24. x = 14, y = 4
Q25. (-2, 1.5) minimum
Q26. x = 2, y = 5
Q27. i) 1992 C = 2597.16 1992 C
t
= 2686.05
ii) 1991 C = 2647.07 1992 C
t
= 2689.71
Q28. H
0
: p
1
= p
2
= p
3
= p
4
= 0.25 H
1
: H
0
not true obs X
2
= 8.2 Dont Rej H
o

Q29. H
0
: p = 0.58 H
1
: p < 0.58 obs Z = -2.14 Rej H
o
p = 0.0162
Q30. a) 0.8770 1.0687 1.1813 0.8730 c) 0.745, 0.7555
d) 0.65, 0.81
Q31. m = 4 n = 2, maximum



14
Q32. i)
26 6 14
16 3 10
21 3 15
|
\

|
.
|
|
|
ii)
27 5 39
3 1 6
9 3 18

|
\

|
.
|
|
|
iii) cant do
iv) 27 v)
0 0 0
4 2 5
20 10 25
|
\

|
.
|
|
|
vi)
5 0 2
1 6 1
1 1 7
|
\

|
.
|
|
|
vii) cant do
viii)

|
\

|
.
|
14 5
36 13
ix)
4
2
5
|
\

|
.
|
|
|
x)

|
\

|
.
|
28 22
4 14


Q33. Ho: p
1
- p
2
= 0 H
1
: p
1
- p
2
0 obs Z = 3.06 Rej H
o


Q34. A = 84.19083, B = 125.96557, C = 9, D = 13.996, E = 6.015
H
0
:
1
= 0 H
1
:
1
0 obs F = 6.015 Rej H
o
at = .05
Q35. k = 127, l = 122 maximum
Q36. a) multiplicative s
i
= 4

b) Q1 Q2 Q3 Q4
1989 75 79.7 85.4 101.0
1990 79.76

c) and d) Q4 1990 T = 84.12 Y = 81.60 Q2 1991 T = 86.48 Y = 106.37
Q37. x = 1, y = 5
Q38. H
0
: p = 0.8 H
1
: p < 0.8 obs Z = -1 Dont Rej H
o

Q39. H
0
: p
1
= 9/16 p
2
= 3/16 p
3
= 3/16 p
4
= 1/16 H
1
: H
0
not true
obs X
2
= 0.45 Dont Rej H
o

Q40. H
0
: Distn of party preference is the same for each state H
1
: distn is different for a t least one
state obs X
2
= 29.9 Rej H
o

Q41. ( )
1
4
1
2
, maximum
Q42. H
0
: income and church attendance independent H
1
: not independent
obs X
2
= 4.494 Dont Rej H
o

Q43. y = 33.314 - 1.128 temp - 0.267 rain R
2
= 0.55
(7.242) (0.304) (0.507)
iii) H
0
:
1
=
2
= 0 H
1
: at least one 0 obs F = 7.445 Rej H
o

iiv) H
0
:
3
= 0 H
1
:
3
< 0 obs t = -3.710 Rej H
o
v) 8.291
Q44. H
0
: p
1
= 0.03 p
2
= 0.10 p
3
= 0.25 p
4
= 0.47 p
5
= 0.15 H
1
: H
0
not true
obs X
2
= 9.528 Dont Rej H
o

Q45. x = 2
1
/
2
y =
1
/
2

Q46. a) A = 42.0954 B= 6.291992 C = 115.65079 D = 7 E = 6.6903
b) H
0
:
1
=
2
= 0 H
1
: H
0
is not true obs F = 6.69 Rej H
o

Q47. x = -1, y = 2
Q48. Xdbar = 4.51, LCL = 3.85, UCL = 5.17
Rbar = 0.9, LCL = 0, UCL = 2.45
Q49. H
0
: age and type of car independent H
1
: not independent obs X
2
= 9.64 Dont Rej H
o

Q50. a) multiplicative s
i
= 4
b) Q1 Q2 Q3 Q4
1993 212.77 218.10 224.76 228.12
1994 224.47

c) and d) Q1 1996 T = 271.37 Y = 255.09 Q3 1998 T = 319.27 Y = 335.23
15
Q51. H
0
: = 0.5 H
1
: > 0.5 obs Z = 1.483 Dont Rej H
o

Q52. H
0
: style and profession independent H
1
: not independent obs X
2
= 7.70 Dont Rej H
o

Q53. H
0
:
1
= 0.8
2
= 0.15
3
= 0.05 H
1
: At least one I is different obs X
2
= 12.8 Rej H
o

Q54. H
0
: = 0.1 H
1
: > 0.1 obs Z = 2.67 Rej H
o
p = 0.0038
Q55. i) A = 2.034 B = -0.583 C = 6758.62 D = 1990.975 E = 3.3948
iii) t* = 2.034 significant iii) H
0
:

= -26H
1
: > -26 obs t = 0.04 Dont Rej H
o

iv) H
0
:
1


= 0 H
1
:
1
0 F* = 3.3948 Dont Rej H
o

Q56. a) Q3 1990 i) 2792.8 ii) 2719.38 iii) 2719.37 iv) 2865.38
b) Q4 1990 I) 2856.95 ii) 2724.12 iii) 2724.12 iv) 2903.34

Q57. i) y = 239.532 + 7.37 age + 10.181 size + 17.309 attend
(108.876) (1.459) (2.273) (9.040) R
2
= 0.92
iii) H
0
:
1
=
2
=
3
= 0 H
1
: at least one 0 obs F = 88.3 Rej H
o

iv) H
0
:
2
= 15 H
1
:
2
< 15 obs t = -2.120 Rej H
o
v) $1526.54

Q58. F* = 3.07, obs. F = 1.02, yes
Q59. pbar = 0.05, LCL = 0.0, UCL = .1534
Q60. c) H
0
:
1
=
2
= .. =
5
= 0 H
1
: H
0
not true obs F =43.29 p value = 0.0000 Rej H
0

y = $6802.40 e) y = $6430.60 f) H
0
:
2
= 0 H
1
:
2
> 0 obs t = -1.407 Reject H
0

g) H
0
:
4
= 0 H
1
:
4
< 0 obs t = -2.945 Reject H
0


Q61. 4 levels, sizes 18, 15, 21, 11; F = 2.95, p = 0.04; means = 226.73, 238.79, 232.58, 239.82