1. Given the data in GDPandProductivity.xls we would like to measure the impact of productivity in manufacturing, and total productivity on GDP.
GDP = Gross Domestic Product billions of dollars
Prod.Manuf.: Productivity of Manufacturing, index 2012=100
Prod.Bus.: Productivity of all Businesses, index 2012=100
a) Deseasonalize the three series
b) Using the deseasonalized series, detrend each of these series
c) Run a regression using GDP as the dependent variable and both productivity indexes as the independent ones.
d) Test for autocorrelation and correct the model if necessary.
e) Comment the results you obtained
2. Given the data in Cons.xls we would like to calculate the following model:
Where C is consumption and Y is total Income.
a) Use the White and the Breusch-Pagan tests for heteroscedasticity. Show your results.
b) If you found heteroscedasticity assume that the variance of the error term is proportional to the explanatory variable. Solve the model and show the results.
c) Now, assume that you do not know the type of variance for the error term. Run the econometric model using different, unknown, variances for the error term.
3. Points) Given the data in Copper.xls we would like to estimate a model for the determinants of the price of Copper in the U.S. from 1951 to 1980.
where: C= 12-month average U.S. domestic price of copper (cents per pound)
G=annual gross national product ($, billions)
I=12-month average index of industrial production
L=12-month average London Metal Exchange price of Copper (pounds sterling)
H=number of housing starts per year (thousands of units)
A=12-month average price of aluminum (cents per pound)
a) Run this model (be careful of the specification of the model).
b) Use the Durbin-Watson statistic and the Breusch-Godfrey tests to test for autocorrelation. Show your results.
c) Assume that the error terms follow an AR(1) process. Run a regression taking care of this AR(1) process. Show the results.