*
* Cointegration example from page 303
*
open data auto2.asc
calendar(q) 1959:1
data(format=prn,org=columns) 1959:1 1992:1 y x2 x3 x4
*
* The case with no deterministic trend in the data and an intercept in the CE is
* represented by the option DETERM=RC (RC means Restricted Constant - the constant
* is restricted to the cointegrating vector, rather than being free in the VAR
* equation). The JohMLE procedure counts the number of lags in the undifferenced
* VAR, while the output shown is based upon 4 lags in the *differenced* VAR. To
* get the results to match up, the LAGS option needs to be set to 5. (This is also
* true if you use CATS).
*
@JohMLE(lags=5,determ=rc) 1959:1 1990:4
# y x2 x3 x4