The Reagan.xls file contains information for the following two variables:  consumption (C) and disposable personal income (YD).  Both variables are in billions of dollars and were measured over the 1959 Q1 to 1995 Q2 period.

I.   EALimdep Programming:

     You will have to create two new variables (namely, REAGAN and YDREAG) from the existing variables in the Excel worksheet.  To do so:

• Start up the EALimdep program.
• Import the Reagan.xls file.
• Create the REAGAN variable by clicking on <Project><New><Variable>, typing in REAGAN in the Name box and IND(85,116) in the Expression box, and then clicking on OK.  Then create the YDREAG variable by clicking on <Project><New><Variable>, typing in YDREAG in the Name box and YD*REAGAN in the Expression box, and then clicking on OK.

II.   Assignment:

a.   What data are contained in the Reagan.xls file?  For what time period?
b.   The REAGAN variable is a (1,0) dummy variable for the Reagan quarters (1980I to 1987IV).  What does the variable specifically indicate?
c.   The YDREAG variable is an interactive term between the YD and REAGAN variables.  What does the variable specifically indicate?
d.   Generate descriptive statistics and rangesfor all of the variables using EALimdep (click here for help).   What do they tell us about the average values, and dispersion of the variables.  Discuss in the specific terms in which each variable is measured.
e.    Run the following regressions:

        i.     C on YD
        ii.    C on YD and REAGAN
        iii.   C on YD and YDREAG
        iv.   C on YD, REAGAN and YDREAG

        Compare and contrast the four regression models.  What's different between them, and what do they imply about what determines consumption?
f.   Interpret the regression coefficients for each model.
g.   Do you believe the Reagan years should have had a specific effect on consumption?  If so, what kind of coefficients would be observed in each model?  Conduct appropriate t-tests on the coefficients for each model.  State your hypotheses.
h.   Interpret the R squared values
i.   Conduct and interpret F-tests on each of the four regression models.
j.   Conduct F-tests between the first and second regressions and the first and third regressions to see if the dummy variables' coefficients differ significantly from zero.  Do the F-test results agree with their corresponding t-tests?
k.   Given:

C = B₁ + B₂YD + B₃REAGAN + B₄YDREAG + e                     (unrestricted model)

C= B₁ + B₂YD + e       (restricted model)

Did the Reagan years alter the consumption-income relationship?  How so?  Conduct an F-test and set up the null and alternative hypotheses.