Time Series in Eviews

Chapter 8 Homework Help

The data set COINT6.XLS contains the three simulated data series – W, Z and Y – used in the text and illustrated in the lecture. 

(a).  The data are plotted in Figure 6.2.  The data do not appear to be stationary.  What sort of non-stationary model(s) would you consider as possible candidates – trend stationary, unit root with drift, or unit root without drift?

(b).  Conduct an augmented Dickey-Fuller test (as below) for each variable, using lag lengths (p) of 0 and 4.  Report your results for the γs  in a table.  What do you conclude?     

Now, impose the null of a unit root and test for the significance of the constant.  What are your results?  Finally, re-test for a unit root but do not include a constant.  What are your results?  Do they match with your assessments in (a)?

(c).  Estimate the following long-run cointegrating relationships:

Report your results for the βs in a table, along with the t-statistics.  Why can’t you conduct inference using these t-statistics? 

(d).  Using the 3 sets of residuals from (c), test for stationarity ( you do not need to report your results).  Why can’t we use the usual DF tables to make inferences about stationarity?  Which table is appropriate for finding the correct critical values?  Using a 5% critical value, what do you conclude?

(e).  Now, estimate an error correction model using the error-term from the 1^st equation in part (c) and 1 lag of Δy, Δz, and Δw.   The easiest way to do this is estimate a VAR model, and enter the equilibrium errors as “exogenous variables”.  Remember to lag them!  Present your results for the adjustment coefficients in a concise table with coefficient estimates and t-statistics.   Which error-correction coefficients are statistically different from zero?  What critical value did you use?  Are there any pitfalls in using that critical value? 

(f) .  Are the estimated error-correction coefficients consistent with LR convergence to your estimated cointegration vectors in (c)?  Why or why not?