... as described by these two wikipedia pages:
http://en.wikipedia.org/wiki/Moving_average http://en.wikipedia.org/wiki/Moving-average_model
In the former, the predicted value for the variable of interest is given by (in the simplest case) the mean of a few lagged values.
In the latter, the predicted value is given by some weighted 'error' terms. Is that error supposed to be the model's prediction subtracted from the actual value that obtained?
Are they similar things or totally different?
