I have been looking into the time series features in Maple and was eager to apply the models to one specific example containing 47 data points (attached).
When I run the ESM routine, Maple provides a forecast based on a (A,N,N) configuration. You will notice that the forecast for the following 12 data points is a constant value. I have also noticed this for several other data set examples and I would have expected the predictions to vary across the next 12 data points.
Does the (A,N,N) configuration in Maple automatically provide an optimal forecast and can anyone advise me on how to specify all possible combinations of (error, trend, season) models?
Thanks you for reading.