Re: Problem with forecast se in the forecast package

Previous Topic Next Topic
classic Classic list List threaded Threaded
1 message Options
Open this post in threaded view
Report Content as Inappropriate

Re: Problem with forecast se in the forecast package

francis pampush
Mr. Shah, you may need to get under the hood to see how the "standard deviations" are actually computed in both of your examples -  - probably starting with the computation of the "pointest.sd" for your forecast example f2.
According to the help menu for the function forecast() from your second example, the outputs "lower" and "upper" refer to prediction intervals (with the default confidence levels of 80 & 95%).  These are not confidence intervals.  You used those PIs to compute standard deviations as though the PIs were confidence intervals.  If you convert the PIs to the related CIs, you will then be able to make the comparison you are trying to make.
If that does not work, you will next want to check to make sure that you actually are making a standard error-to-standard error (i.e., sd/sqrt(N) for the se of sampling distribution of means).  You don't want to compare a standard deviation to a standard error!
Best wishes.
Frank Pampush

        [[alternative HTML version deleted]]

[hidden email] mailing list
-- Subscriber-posting only. If you want to post, subscribe first.
-- Also note that this is not the r-help list where general R questions should go.