Wensui Liu
2016-Jan-03 22:57 UTC
[R] How to calculate the prediction interval without knowing the functional form
If I have predictions derived empirically without knowing the functional form, is there a way to calculate the prediction interval? Thanks -- WenSui Liu https://statcompute.wordpress.com/ [[alternative HTML version deleted]]
Bert Gunter
2016-Jan-03 23:05 UTC
[R] How to calculate the prediction interval without knowing the functional form
Standard answer: bootstrap. However, "derived empirically" is too vague to know whether the standard answer applies. As this appears to be primarily a statistics, not an R issue, I suggest that you post on a statistics list like stats.stackexchange.com instead, perhaps with some more details on what "derived empirically" means. Cheers, Bert Bert Gunter "The trouble with having an open mind is that people keep coming along and sticking things into it." -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip ) On Sun, Jan 3, 2016 at 2:57 PM, Wensui Liu <liuwensui at gmail.com> wrote:> If I have predictions derived empirically without knowing the functional > form, is there a way to calculate the prediction interval? > > Thanks > > > -- > WenSui Liu > https://statcompute.wordpress.com/ > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.
Duncan Murdoch
2016-Jan-03 23:21 UTC
[R] How to calculate the prediction interval without knowing the functional form
On 03/01/2016 5:57 PM, Wensui Liu wrote:> If I have predictions derived empirically without knowing the functional > form, is there a way to calculate the prediction interval?As Bert said, this isn't really an R question, but I'd say the answer is "probably not". The prediction interval depends on the uncertainty of individual observations, and that isn't usually reflected in predictions. For example, if y_i is N(mu, sigma^2), and we fit the model to N observations, all of our predictions will be ybar, and there will be no indication of sigma. Duncan Murdoch