dear members,
I am considering whether to set biasadj = TRUE in
auto.arima in forecast package. If TRUE, the mean is very nearly equal to the
original data. So there is no point in doing the transformation at all...
If FALSE, the point forecast is the median which is different from the mean.
So if I have to get more accuracy and gain some thing by the BoxCox
transformation I have to set the biasadj argument to FALSE.
Am i right?
PLease flag me if this question doesn't belong to R proper and doesn't
fit to this list.
THanking you,
Yours sincerely,
AKSHAY M KULKARNI
[[alternative HTML version deleted]]