[R] prediction interval for new value

[Michael Grant]

Sachin,

One of the references in the link is a EPA groundwater statistics training
manual from the early nineties. For a more complete related discussion see pp.
50-62 of the July 1992 EPA draft addendum on groundwater monitoring statistics
at

http://www.epa.gov/correctiveaction/resource/guidance/sitechar/gwstats/gwstats.htm
 

Regarding discussion the prediction interval for linear regression you might
want to read sections of chapter 9 in Helsel and Hirsch:

http://pubs.usgs.gov/twri/twri4a3/

I would also recommend working of the examples in these or other references
with R in order to 1.) reassure yourself that the code is doing what you want
and 2.) you are comfortable with the concepts. I seem to recall seeing the
terms 'tolerance interval' and 'prediction interval' mingled in some
discussions--shades of Peter's 'hardly universal' comment. This means that
there is even more incentive to develop some facility with the concept and code
by working a few out first.

HTH


Regards,
Michael Grant

--- Peter Dalgaard <p.dalgaard at biostat.ku.dk> wrote:

> Sachin J <sachinj.2006 at yahoo.com> writes:
> 
> > Google search gave me this: 
> >    
> >  
> http://ewr.cee.vt.edu/environmental/teach/smprimer/intervals/interval.html
> >    
> >   TIA
> >   Sachin
> 
> With those definitions (which are hardly universal), tolerance
> intervals are the same as prediction intervals with k == m == 1, which
> is what R provides.
> 
>    
> > 
> > Peter Dalgaard <p.dalgaard at biostat.ku.dk> wrote:
> >   Sachin J writes:
> > 
> > > RUsers: 
> > > 
> > > Just confirming, does predict function with interval="prediction"
> > > option gives prediction interval or tolerance interval?. Sorry for
> > > reposting this question.
> > 
> > 
> > Is there any definition of tolerance interval that is different from
> > prediction interval?
> > 
> > (Tolerance intervals in the medical sense means intervals that are
> > designed to detect patients with abnormal levels of serum cholesterol
> > (say).)
> > 
> > -- 
> > O__ ---- Peter Dalgaard Øster Farimagsgade 5, Entr.B
> > c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K
> > (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918
> > ~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk) FAX: (+45) 35327907
> > 
> > 
> >  		
> > ---------------------------------
> > Do you Yahoo!?

> 
> -- 
>    O__  ---- Peter Dalgaard             Øster Farimagsgade 5, Entr.B
>   c/ /'_ --- Dept. of Biostatistics     PO Box 2099, 1014 Cph. K
>  (*) \(*) -- University of Copenhagen   Denmark          Ph:  (+45) 35327918
> ~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)                  FAX: (+45) 35327907
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> 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.
>