Beruflich Dokumente
Kultur Dokumente
Monash University
Semester One 2008 Examination Period
Faculty of Business and Economics
EXAM CODES:
TITLE OF PAPER:
EXAM DURATION:
READING TIME:
10 minutes
Open Learning
Sth Africa
During an exam, you must not have in your possession, a book, notes, paper, calculator, pencil case, mobile
phone or other material/item which has not been authorised for the exam or specifically permitted as noted
below. Any material or item on your desk, chair or person will be deemed to be in your possession. You are
reminded that possession of unauthorised materials in an exam is a discipline offence under Monash Statute 4.1.
AUTHORISED MATERIALS
CALCULATORS
X YES
NO
OPEN BOOK
YES
X NO
YES
X NO
INSTRUCTIONS TO CANDIDATES
Answer ALL questions in this examination paper
Paper is worth 100 marks in total
Page 1 of 10
INSTRUCTIONS TO CANDIDATES
Answer ALL questions in this examination paper
Paper is worth 100 marks in total
Background
As an international economist advising Australias Minister for Trade, you are interested in why
Australia trades more with certain countries and not others. You have data on how much trade takes
place between Australia and various other countries around the globe, and wish to analyse it.
Trade ($ billion)
Mean
Standard Error
Median
Mode
Standard Deviation
Sample Variance
Kurtosis
Skewness
Range
Minimum
Maximum
Sum
Count
205.1348
45.6170
13.4438
3.0797
618.7793
382,887.7697
30.8781
5.1089
5,222.6093
0.0007
5,222.6100
37,744.7985
184.0000
What do you learn about the volume of trade from Australia to the countries in the sample? Be
sure to interpret all the relevant summary statistics in the output.
(6 marks)
Page 2 of 10
(2) Next you use Excel to run a regression of trade volume on a variable taking a value of 1 for all
observations.
SUMMARY OUTPUT
Regression Statistics
Multiple R
0.315447
R Square
0.099507
Adjusted R Square
0.094043
Standard Error
618.7793
Observations
184
ANOVA
df
Regression
Residual
Total
Intercept
X
1
183
184
SS
MS
F
Significance F
7742771 7742771 20.22204
1.23E-05
70068462 382887.8
77811233
Lower 95%
Upper 95%
#N/A
#N/A
115.1319
295.1377
(a) How does the estimated coefficient for the X variable in this regression relate to your
summary statistics in (1)? Interpret the value of this coefficient.
(3 marks)
(b) What do the 95% Confidence Lower and 95% Confidence Upper values for the X
variable tell us? Interpret these values.
(4 marks)
Trade Volume
($ billion)
Less than 100
100-400
More than 400
Total
Agreement
127
20
17
164
No Agreement
15
3
3
21
Total
142
23
20
185
(a) What is the probability a randomly selected country has a trade agreement with Australia?
(2 marks)
(b) What proportion of countries traded up to $400 billion with Australia?
(2 marks)
Page 3 of 10
(c) What is the probability that a randomly selected country which has a trade agreement with
Australia traded between $100 and $400 billion, inclusive?
(3 marks)
(d) What is the probability that a randomly selected country traded between $100 and $400
billion, inclusive, and had a trade agreement with Australia?
(3 marks)
(e) Check whether trade volume is independent of having a trade agreement.
(3 marks)
AGREEMENT
SUMMARY OUTPUT
Regression Statistics
Multiple R
0.1647
R Square
0.027126
Adjusted R Square
0.021781
Standard Error
612.0035
Observations
184
ANOVA
df
Regression
Residual
Total
Intercept
Agreement
1
182
183
SS
MS
F
Significance F
1900678 1900678 5.074589
0.025473
68167783 374548.3
70068462
Page 4 of 10
(a) Give a point prediction for the amount of trade that takes place between Australia and a
country WITHOUT a trade agreement.
(2 marks)
(b) Give a point prediction for the amount of trade that takes place between Australia and a
country WITH a trade agreement.
(2 marks)
(c) Perform a test to determine whether there is a difference in mean trade volume depending
on whether the country has a trade agreement with Australia. Use a critical value approach
the 5%, 1-sided critical value you need is 1.653.
(5 marks)
Regression Analysis
(5) You next estimate a regression model in Excel using the following variables:
TRADE
AGREEMENT
LANDLOCKED =
ENGLISH
DISTANCE
DISTANCE2
Distance squared
Region dummies:
AMERICAS:
ASIA / PACIFIC:
EUROPE:
MIDDLE EAST / AFRICA: 1 if the partner country is located in the Middle East / Africa
region, 0 otherwise
N.B.
SUMMARY OUTPUT
Regression Statistics
Multiple R
0.475683
R Square
0.226275
Adjusted R Square
0.190904
Standard Error
556.5904
Observations
184
ANOVA
df
Regression
Residual
Total
Intercept
Agreement
Landlocked
English
Distance
Distance2
Asia / Pacific
Europe
Middle East / Africa
8
175
183
SS
MS
F
Significance F
15854713 1981839 6.397304
2.78E-07
54213749 309792.9
70068462
t Stat
3.008493
0.52716
-1.33504
0.393889
-2.44603
1.977465
2.190891
0.77397
1.017732
P-value
Lower 95% Upper 95%
0.003013
380.4768 1831.685
0.59875
-187.513 324.1905
0.183597
-374.151 72.21125
0.694142
-143.683 215.3352
0.015433
-0.47855
-0.05115
0.049559
2.93E-08
3E-05
0.029781
36.79227 705.1841
0.439993
-176.291 403.7642
0.31021
-122.843 384.4254
(a)
Interpret each of the estimated coefficients in the model, except the coefficients for
Distance and Distance2. Are the signs what you would expect? Explain.
(12 marks)
(b)
(i) Perform a significance test on the Agreement dummy (use a p-value approach).
(3 marks)
(ii) Contrast the estimate and p-value for the Agreement dummy for this regression with
that in (4). Explain.
(3 marks)
(c)
Roughly sketch a graph of the estimated relationship between Trade and Distance. Label
the axes and the value of Distance where the turning point occurs. Does this relationship
seem reasonable?
(4 marks)
Page 6 of 10
(d)
(i) Perform significance tests on each of the region dummies using a critical value
approach Excel output follows.
(9 marks)
(ii) How is it possible that a test may reject a true null hypothesis? If you were to
increase the significance level of your test, what probability are you increasing? If
you specify a null hypothesis that is actually false, would you expect to reject the null
more or less often when you increase the significance level? Explain.
(5 marks)
Page 7 of 10
(e)
Residuals
3000
2000
1000
0
0
2000
4000
6000
8000
10000
12000
-1000
-2000
Distance
(2 marks)
(iv) Based on your answers to (i)-(iii), do you think this model fits the data well?
(1 mark)
Page 8 of 10
(a)
Mar-05
Mar-04
Mar-03
Mar-02
Mar-01
Mar-00
Mar-99
Mar-98
Mar-97
Mar-96
Mar-95
Mar-94
Mar-93
Mar-92
Mar-91
Mar-90
(b)
By looking at this graph, it would seem that Australias export volume has grown
substantially over the last 15 years in 1990 it was $12.2 billion, growing to $37.7 billion
by 2005. What is the problem with this comparison? Suggest a divisor that would enable
you to make a more appropriate comparison between 1990 and 2005.
(3 marks)
Page 9 of 10
(c)
You are interested in forecasting Australias export volume into the future. You fit a
linear trend model with quarterly seasonal dummies.
N.B. The time trend variable (t) is set to zero in the 4th quarter of 1989.
SUMMARY OUTPUT
Regression Statistics
Multiple R
0.958692
R Square
0.919091
Adjusted R Square
0.913605
Standard Error
2.045236
Observations
64
ANOVA
df
Regression
Residual
Total
Intercept
t
Quarter 2
Quarter 3
Quarter 4
4
59
63
SS
MS
F
Significance F
2803.494 700.8735 167.5532
1.72E-31
246.7964 4.182991
3050.29
t Stat
14.06616
25.47577
1.866923
2.09439
3.029892
P-value
Lower 95% Upper 95%
1.67E-20
8.059068 10.73224
1.57E-33
0.325473
0.38096
0.066883
-0.09697 2.797407
0.04053
0.067583
2.96355
0.003628
0.745225 3.643849
(d)
Carefully interpret all of the estimated coefficients in this model. Do the estimates seem
plausible? Explain.
(10 marks)
(e)
Use the model to forecast export volume for each of the four quarters of 2006 (N.B. t =
65 in the 1st quarter of 2006). Be sure to show your working.
(6 marks)
End of Examination
Page 10 of 10