2. A researcher wishes to model consumption in country Z.
Use the EViews Output in the appendix {Real consumption in $ [CONS], real income in $ [INC] and LN=Natural Log} [60 points]
1. Suppose that researcher, initially, wishes to estimate Keynesian consumption function (KCF). Write down the PRF.
Determine the estimated KCF from the relevant EViews output. [Also provide R2, t-ratios and the F-statistic and the corresponding p-values]
Do the results in part (b) concur with the a priori expectations? Interpret the estimated slope coefficient.
Test for autocorrelation [AC].
Test whether the real consumption [CONS] and income [INC] are stationary using unit root (ADF) tests. [Also determine whether CONS and INC are I(1) or something else]
Test for cointegration between the two series [CONS & INC] using the Engle–Granger (E-G) methodology. Would you say that the regression equation stated in (b) is spurious? Why? Why not?
Comment on the validity of the application of the E-G Approach in (f).
Determine the SRF for the following long-run or equilibrium relationship: LNCONSt =0 +1LNINCt+ut. Is the sign of the slope coefficient of the independent variable as one would expect? Explain briefly. Also interpret the estimated slope coefficient.
Test for cointegration between LNCONS & LNINC using the Engle–Granger methodology. Would you say that the regression equation stated in (h) is spurious? Why? Why not? {Assume that LNCONS & LNINC are I(1)}
Which consumption model among the ones given above would you choose by considering the available information, and why?