[R] building a regression model

[Rui Barradas]

Hello,

The "error" is just a warning, these are not the same thing. And the 
warning says that you're giving a wrong scope argument, one that 
includes the response. See ?add1. Argument 'scope' is "a formula giving 
the terms to be considered for adding or dropping." And you're passing 
the entire data.frame, not just the regressors.
To get rid of the warning, try the following.

add1(mydata.lm, ~ A + B + C + D, test='F')

Hope this helps,

Rui Barradas
Em 04-01-2013 23:53, dada escreveu:
> Hi
>
> I am trying to build a regression model. My data looks like this:
>
> A	B	C	D	        E
> 1	1	1	2.57	5
> 2	0	0	1.64	3
> 0	5	1	4.8	        1
> 1	3	0	3.56	168
> 1	1	1	2.13	1
> 0	3	1	5	        168
> 2	0	0	7.75	28
> 4	0	2	2.85	168
> 3	0	1	1.89	6
> 1	1	1	2.33	3
> 3	2	2	1.77	168
> 1	0	0	1.38	0.04
> 0	6	1	4.57	168
> 2	2	2	2.86	1
> 3	1	1	4.11	168
> 3	0	2	3	         84
> 1	1	1	2.5	         56
>
> where E is a response variable and A, B, C, D are predictor variables. Below
> are commands which I enter in R:
>
> str (mydata)
> mydata.lm = lm (E ~ 1, data = mydata) # to create a blank model
> add1(mydata.lm, mydata, test='F') # to add the best predictor variable to
> the blank model
>
> However at this point the message appears:
>
> Warning messages:
> 1: In model.matrix.default(Terms, m, contrasts.arg = object$contrasts) :
>    the response appeared on the right-hand side and was dropped
> 2: In model.matrix.default(Terms, m, contrasts.arg = object$contrasts) :
>    problem with term 4 in model.matrix: no columns are assigned
>
> What does it mean ? Could you please help ?
>
>
>
>
>
>
> --
> View this message in context: http://r.789695.n4.nabble.com/building-a-regression-model-tp4654701.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at r-project.org 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.