[R] problem with gls finding model terms without specifying data=named.object

Paul.Rustomji at csiro.au Paul.Rustomji at csiro.au
Mon Feb 16 11:41:11 CET 2009


Hello R-help

I am having trouble getting gls to find the R objects that comprise a linear model when the data=named.object option(option!) is not specified.  In the gls() help it states data is "an optional data frame containing the variables named in model, correlation, weights, and subset. By default the variables are taken from the environment from which gls is called".


An example:

> temp <- data.frame(x=1:10,y=11:20+rnorm(10))
> temp
    x        y
1   1 11.52458
2   2 10.77643
3   3 12.56845
4   4 14.48822
5   5 13.58116
6   6 16.26223
7   7 17.89619
8   8 19.40359
9   9 18.56699
10 10 21.05374
> gls(temp$y~temp$x)
Error in eval(expr, envir, enclos) : object "y" not found
> gls(y~x,data=temp)
Generalized least squares fit by REML
  Model: y ~ x 
  Data: temp 
  Log-restricted-likelihood: -14.00387

Coefficients:
(Intercept)           x 
   9.366256    1.135619 

Degrees of freedom: 10 total; 8 residual
Residual standard error: 0.9156084 

 I'm trying hard to avoid having to specify the data option if at all possible.

Paul

-------------------------------

R version 2.8.0 (2008-10-20) (yes I know its not 2.8.1 but nothing in the 2.8.1 news seemed to be relevant, I also tried on a different PC with R 2.7.0 but same problem)

i386-pc-mingw32 

locale:

LC_COLLATE=English_Australia.1252;LC_CTYPE=English_Australia.1252;LC_MONETARY=English_Australia.1252;LC_NUMERIC=C;LC_TIME=English_Australia.1252



attached base packages:

[1] stats     graphics  grDevices utils     datasets  methods   base     



other attached packages:

[1] nlme_3.1-89



loaded via a namespace (and not attached):

[1] grid_2.8.0      lattice_0.17-17

-----------------------------------------------



                Information on package 'nlme'



Description:



Package:       nlme

Version:       3.1-89

Date:          2008-06-07

Priority:      recommended

Title:         Linear and Nonlinear Mixed Effects Models

Author:        Jose Pinheiro <Jose.Pinheiro at pharma.novartis.com>, Douglas Bates <bates at stat.wisc.edu>,

               Saikat DebRoy <saikat at stat.wisc.edu>, Deepayan Sarkar <Deepayan.Sarkar at R-project.org>,

               the R Core team.

Maintainer:    R-core <R-core at R-project.org>

Description:   Fit and compare Gaussian linear and nonlinear mixed-effects models.


Paul Rustomji
Rivers and Estuaries
CSIRO Land and Water
GPO Box 1666
Canberra ACT 2601

ph +61 2 6246 5810
mobile 0406 375 739




More information about the R-help mailing list