Bug? (PR#1710)
ripley@stats.ox.ac.uk
ripley@stats.ox.ac.uk
Wed, 26 Jun 2002 12:34:21 +0100 (BST)
This is NOT A BUG.
If you want a squared term, you need to include I(carb^2). See ?formula
or even `An Introduction to R'.
Please
1) Use an informative subject line, as only the subject appears in the
BUGS summary file.
2) Read the Bugs section in the FAQ, and only post to R-bugs if you
are absolutely sure you understand.
On Wed, 26 Jun 2002 Lauri.Mehtatalo@metla.fi wrote:
> Hi,
>
> I tried to do a multiple linear model from the example
> dataset Formaldehyde. However, the function lm() did not
> estimate the coefficient of the term carb^2. The same
> problem occurred with the (nlme)dataset Pixel with both
> function lme() and lm(). I am using the windows version of
> R 1.5.1
>
> Lauri Mehtatalo
>
> The Formaldehyde example:
> > data(Formaldehyde)
> > lm(optden~carb+carb^2,data=Formaldehyde)
>
> Call:
> lm(formula = optden ~ carb + carb^2, data =
> Formaldehyde)
>
> Coefficients:
> (Intercept) carb
> 0.005086 0.876286
>
> The Pixel example:
>
> > {pkg <- select.list(sort(.packages(all.available = TRUE)))
> + if(nchar(pkg)) library(pkg, character.only=TRUE)}
> Loading required package: nls
> Loading required package: lattice
> > data(Pixel)
> >coef(lme(pixel~day+day^2,data=Pixel,random=list(Dog=
> ~day,Side=~1)))
> (Intercept) day
> 1/L 1048.995 0.32564928
> 1/R 1043.659 0.32564928
> 10/L 1144.898 -0.63867455
> 10/R 1099.883 -0.63867455
> 2/L 1054.011 0.65786288
> 2/R 1048.726 0.65786288
> 3/L 1065.000 0.45552014
> 3/R 1075.759 0.45552014
> 4/L 1088.173 0.93300591
> 4/R 1088.854 0.93300591
> 5/L 1125.918 -0.91429707
> 5/R 1129.309 -0.91429707
> 6/L 1112.721 -0.72735611
> 6/R 1094.164 -0.72735611
> 7/L 1133.443 -1.69593549
> 7/R 1156.943 -1.69593549
> 8/L 1101.205 0.05435070
> 8/R 1087.689 0.05435070
> 9/L 1093.513 0.05987474
> 9/R 1071.482 0.05987474
> > coef(lm(pixel~day+day^2,data=Pixel))
> (Intercept) day
> 1076.820678 1.341873
> - - - - - - - - - - - - -
> Lauri Mehtätalo
> Metsäntutkimuslaitos
> Joensuun tutkimuskeskus
> PL 68 (Yliopistokatu 7)
> 80101 JOENSUU
> lauri.mehtatalo@metla.fi
> p. 013-251 4110
> - - - - - - - - - - - - -
>
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
> r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
> Send "info", "help", or "[un]subscribe"
> (in the "body", not the subject !) To: r-devel-request@stat.math.ethz.ch
> _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
>
--
Brian D. Ripley, ripley@stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272860 (secr)
Oxford OX1 3TG, UK Fax: +44 1865 272595
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._