R help - Feb 2007 - Access to column names stored in a vector in lm procedure

Hello everybody

I have to run many statistical tests that are identical, with the
exception of the dependent variable. Is there a possibility to store the
dependent variable names e.g. in a vector (in the below mentioned example 
called “variable”) and to use the content of this vector in a simple
statistical test (e.g. a regression). I would like to write the
statistical procedure only once…


For example: I would like to store 100 dependent variable names in a
vector called “variable” and then I would like to do a simple regression
with each of theses variables and the variable “length” using e.g. the
while function. I would then extract e.g. the t-value and add it to a
vector (“result”) that contains all results. Something like that:



“variable” should contain the names of the 100 dependent variables (Var1, 
Var2, … Var100)

while(i<101){
result<-c(result,coef(summary(lm(variable[i] ~ length, data = data2)))[2,4]);
i<-i+1
}

This example does not work since the lm function does not recognize the
dependent variable’s name.

Does somebody know how to store the names of the dependent variables in
e.g. a vector and to make them available for the lm function?

Many thanks

Patrick

You can regress multiple dependent variables at once by placing them
in a matrix.   For example, this regresses each of the first 4 columns of
the builtin iris dataset against Species, the last column:

iris4 <- as.matrix(iris[,-5])
summary(lm(iris4 ~ Species, iris))


On 2/2/07, mcnpf748 at mncn.csic.es <mcnpf748 at mncn.csic.es>
wrote:
> Hello everybody
>
> I have to run many statistical tests that are identical, with the
> exception of the dependent variable. Is there a possibility to store the
> dependent variable names e.g. in a vector (in the below mentioned example
> called "variable") and to use the content of this vector in a
simple
> statistical test (e.g. a regression). I would like to write the
> statistical procedure only once?
>
>
> For example: I would like to store 100 dependent variable names in a
> vector called "variable" and then I would like to do a simple
regression
> with each of theses variables and the variable "length" using
e.g. the
> while function. I would then extract e.g. the t-value and add it to a
> vector ("result") that contains all results. Something like that:
>
>
>
> "variable" should contain the names of the 100 dependent
variables (Var1,
> Var2, ? Var100)
>
> while(i<101){
> result<-c(result,coef(summary(lm(variable[i] ~ length, data =
data2)))[2,4]);
> i<-i+1
> }
>
> This example does not work since the lm function does not recognize the
> dependent variable's name.
>
> Does somebody know how to store the names of the dependent variables in
> e.g. a vector and to make them available for the lm function?
>
> Many thanks
>
> Patrick
>
>
>
> ______________________________________________
> 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.
>
>
>