This preview shows pages 1–3. Sign up to view the full content.
Key  Exam 2A
Please use the data set
Exam2 – Demand for Money.xls
for all analyses. Variables in the Excel spreadsheet
are for the years 19601983 and are defined as follows:
±
GNP –
the U.S. Gross National Product in billions of dollars.
±
M1 – U.S. demand for money in billions of dollars.
±
TBillR – Treasury Bill Rate in percent.
1. Before you do anything, save the Excel spreadsheet in the
Dropbox
as “
yourlastname Exam2.xls
.”
Then
click the “Save” icon often as you work. We have had crashes in the lab, and we are under time constraints.
2. These data will be used to estimate a demand function for money for the U.S. The demand for money
should be affected by both the cost of holding money (a
price of money
– TBill Rates) and a measure of
income for the economy (GNP):
01
2
tt
t
t
M
1T
B
i
l
l
R
G
N
P
u
β
ββ
=+
+
+
(5)
a. Every regression model has certain parts. For the model above, explain/name each of the elements (i.e.,
what are:
M1,
the
β
s, TBillR
, and
u
?)
M1
is the
dependent variable
.
TBillR
and
GNP
are the
independent variables
– they cause the dependent variable to change.
The
β
s
are the
population parameters
– the constants that define how the independent variables
affect the dependent variable.
u
is the
disturbance, the random component of the model
.
(4)
b. Before completing any analyses, what
expectations do you have for
β
1
. How is this used in a hypothesis
test.
This is a demand function – the demand for money – and TBill Rate is used to measure the
price
of
holding money, so
we
expect the effect
β
1
to be negative
.
This expectation,
β
1
< 0
, is used for the alternative hypothesis.
(6)
3.
In Excel
,
create a correlation matrix/table
for the three variables, Money Demand, TBill Rates and GNP.
Briefly
interpret each correlation coefficient
.
There are
very strong positive correlations among all variables
. We can see that GNP and M1
are nearly perfectly positively correlated. M1and TBillR also has a very strong positive correlation of
0.856 as do the variables TBillR and GNP (correlation of 0.874).
These strong positive correlations
among the independent variables suggest that omitting one of them, if they were important, would
result in specification bias (more later).
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document(6)
4. Use
Excel functions
This is the end of the preview. Sign up
to
access the rest of the document.
 Winter '10
 DanielLass
 Gross National Product

Click to edit the document details