an indirect solution is to use the following:
pro - function(indep, dep){
d - data.frame(indep)
form - formula(lm(dep~., data=d))
assign(d, d, envir=.GlobalEnv)
res - step(lm(dep~X1, data=d), scope=form, trace=0,
direction=f)
rm(d, envir=.GlobalEnv)
res
}
#
m - rnorm(100*5); dim(m) - c(100, 5); colnames(m) - paste(X, 1:5,
sep=)
Y - as.vector(1 + m[, sample(5, 2)]%*%rnorm(2))
pro(m, Y)
step(lm(Y~X1, data=data.frame(m)), scope=Y~X1+X2+X3+X4+X5, trace=0,
direction=f)
I hope it helps.
Best,
Dimitris
Dimitris Rizopoulos
Ph.D. Student
Biostatistical Centre
School of Public Health
Catholic University of Leuven
Address: Kapucijnenvoer 35, Leuven, Belgium
Tel: +32/16/336899
Fax: +32/16/337015
Web: http://www.med.kuleuven.ac.be/biostat/
http://www.student.kuleuven.ac.be/~m0390867/dimitris.htm
- Original Message -
From: vasilis pappas [EMAIL PROTECTED]
To: r-help@stat.math.ethz.ch
Sent: Friday, April 01, 2005 9:29 PM
Subject: [R] step error
Could anyone tell me what am I doing wrong?
pro-function(indep,dep){
+ d-data.frame(indep)
+ form-formula(lm(dep~.,data=d))
+
forward-step(lm(dep~X1,data=d),scope=form,trace=0,direction='f')
+ return(forward)
+ }
pro(m,q)
Error in inherits(x, data.frame) : Object d not
found
Where q is a vector with the dependent variable's
values
and m is a matrix containing the values of the
independent variables.
While writing the above without a function form has no
problem, that is :
d-data.frame(m)
form-formula(lm(q~.,data=d))
forward-step(lm(q~X1,data=d),scope=form,trace=0,direction='f')
forward
Call:
lm(formula = q ~ X1 + X2 + X5, data = d)
Coefficients:
(Intercept) X1 X2 X5
-15.7988.7656.774 -4.245
Thank you in advance!
Vasilis
__
R-help@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
__
R-help@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
