The data for the homework assignment is available in
EasyReg space delimited text format (*) and in
Excel CSV format (**).
This data file contains seasonally adjusted quarterly time series for the US:
- %DIF1[Real consumption] = Percentage change in real consumption
- %DIF1[Real GDP] = Percentage change in real GDP
- %DIF1[Real investment] = Percentage change in real investment
Let y(t) = (y(1,t), y(2,t), y(3,t) )', where
- y(1,t) = %DIF1[Real investment]
- y(2,t) = %DIF1[Real GDP]
- y(3,t) = %DIF1[Real consumption]
- Determine whether you need to include a time trend in the VAR for y(t).
- Determine the lag length p of the VAR(p) model for
y(t), using a sequence of Wald tests, starting from p = 6.
- Test whether the changes in real consumption and GDP do NOT Granger-cause the change
in investment. If so, implement the Granger causality restriction involved in your VAR.
- Plot the response of %DIF1[Real GDP] and %DIF1[Real consumption]
to a unit shock in the innovation of %DIF1[Real investment] over a horizon of 12 quarters.
- Specify and estimate a structural VAR model which assumes that the nonstructural error
of the change in real consumption is determined by the nonstructural error of the change in
real GDP, and the latter is determined by the nonstructural error of the change in investment,
whereas the change in real investment is exogenous.
Is this structural model supported by the data?
(*) Click on the link, and save the file as a text file (with extention *.TXT) on your hard disk via the
"Save as .." option of your web browser.
(**) If you use Internet Explorer and you click on the link the file will be
automatically imported in Excel, provided of course that you have installed Excel
on your computer. Then use the "Save as .." option of Excel to save
the file on your hard disk. Netscape and Mozilla Firefox will display and save the file
as a text file. Therefore, if you use Netscape or Mozilla Firefox, follow the
procedure under (*), and rename the file as INPUT2.CSV if it is saved with extention ".TXT".