*
* Example 6.4.1 from page 202
*
open data dowj.dat
data(format=free,org=columns) 1 78 dowj
*
* In order to forecast the original (undifferenced) series easily, the DIFFS
* option is used on the BOXJENK instruction. We also use the DEFINE option to
* define a forecasting equation.
*
boxjenk(diffs=1,ar=1,demean,maxl,define=doweq) dowj
*
* The numbering scheme for the forecasts in the text is going to be slightly
* different from the one used here. With 78 data points, the first two
* out-of-sample periods will be 79 and 80.
*
uforecast(equation=doweq,stderrs=ferr) forecast 79 80
*
* This prints the forecasts and the standard errors (square root of the variance).
*
print / forecast ferr