Skip to content

inference for customized regression in R?

3 messages · R. Michael Weylandt, Michael

#
Given the caveats Ted describes here:
http://tolstoy.newcastle.edu.au/R/help/05/06/5992.html it seems that
bootstrapping might be the only way to get (somewhat) credible
prediction intervals: the boot package on CRAN can help to facilitate
getting these. Here's some documentation for CI:

http://www.statmethods.net/advstats/bootstrapping.html

but you'll need to adopt it for a prediction interval, which might
entail hacking boot.ci().

You might also see if this question, by someone who most certainly
isn't you because cross-posting is discouraged, gets some answers:
http://stats.stackexchange.com/questions/26277/how-to-bootstrap-prediction-intervals-for-customized-regression-models-in-r

Michael Weylandt
On Wed, Apr 11, 2012 at 10:29 AM, Michael <comtech.usa at gmail.com> wrote: