ECON2206/ECON3290: Introductory Econometrics
Session 1, 2010
Course Project Solution Guide
Each question is worth 1 mark  and there are 20 questions in total. Note Instruction (d): ‘Remember that
when performing statistical tests, always state the null and alternative hypotheses, the test statistic and it’s
distribution under the null hypothesis, the level of signi
fi
cance and the conclusion of the test.’ Full credit
cannot be given if this is ignored. A printout of the SHAZAM output must be attached at the end of the
answers (otherwise 5 marks are to be deducted).
(1)
What is the average, minimum, and maximum value and standard deviation for each of the variables in
the HPRICEN.RAW sample ?
23926
2
5394
5
4474
6
3615
3
8835
37
756
5000
0
006
3
850
3
560
1
130
18
70
50001
88
976
8
710
8
780
12
130
71
100
9414
8
7
1497
1
1699
0
7322
2
1206
14
916
(2)
Would you expect the correlation between
and
to be positive or negative ? Would you expect
the correlation between
and
to be positive or negative ? Explain.
What is the correlation between
and
and
and
, in the sample ?
I would expect the correlation between
and
to be negative  that prices will be higher in areas
where there is less pollution or lower
re
fl
ecting consumer’s demand. People would be willingness to pay
more for cleaner air and a nice environment, which in turn implies a negative association between
and
 which is a measure of air pollution. By similiar reasoning, I would also expect a negative association
between
and
 which is a local area ’bad.’ High crime areas are undesirable, so people would be
willing to buy houses in higher crime areas only if the price is lower.
From the sample of data, the raw correlations are
(
)
=
−
0
35732
, and
(
)
=
−
0
34937
.
Simple Regression Model
(3)
Consider the simple regression model:
log(
) =
0
+
1
log(
) +
(1)
What is the in
terpretation of the coe
ﬃ
cient
1
in the model ?
What is the interpretation of the coe
ﬃ
cient
in (1) ?
Explain.
The coe
ﬃ
cient
1
represents the change in expected
log(
)
due to a one unit change in
log(
)
. Given
the loglog functional form,
1
is also the elasticity of expected
with respect to
.
The coe
ﬃ
cient
0
is expected
log(
)
when
log(
)
is equal to 0 [note: it is not
d
log(
)
when
= 0
].
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview.
Sign up
to
access the rest of the document.
 Three '11
 yang
 Econometrics

Click to edit the document details