Hi,

The step-by-step guide to ARIMA modelling instructs one to derive a (possible) MA coefficient from the Yule-Walker equations if one is using Excel. My question is: if the software we're using does work out the MA coefficient automatically (and presumably more accurately), can we use the package's estimate for the coefficient, or must we absolutely derive our estimate from the Yule-Walker equations? 

[NEAS: If your software computes the MA coefficient, that is best; don't bother with the Yule Walker estimate.]