New to R; please excuse me if this is a dumb question. I tried to RTFM; didn't help. I want to do a series of regressions over the columns in a data.frame, systematically varying the response variable and the the terms; and not necessarily including all the non-response columns. In my case, the columns are time series. I don't know if that makes a difference; it does mean I have to call lag() to offset non-response terms. I can not assume a specific number of columns in the data.frame; might be 3, might be 20. My central problem is that the formula given to lm() is different each time. For example, say a data.frame had columns with the following headings: height, weight, BP (blood pressure), and Cals (calorie intake per time frame). In that case, I'd need something like the following: lm(height ~ weight + BP + Cals) lm(height ~ weight + BP) lm(height ~ weight + Cals) lm(height ~ BP + Cals) lm(weight ~ height + BP) lm(weight ~ height + Cals) etc. In general, I'll have to read the header to get the argument labels. Do I have to write several functions, each taking a different number of arguments? I'd like to construct a string or list representing the varialbes in the formula and apply lm(), so to say [I'm mainly a Lisp programmer where that part would be very simple. Anyone have a Lisp API for R? :-}] Thanks, chris Chris Elsaesser, PhD Principal Scientist, Machine Learning SPADAC Inc. 7921 Jones Branch Dr. Suite 600 McLean, VA 22102 703.371.7301 (m) 703.637.9421 (o)
Try this:
lm(Sepal.Length ~., iris[1:3])
# or
cn <- c("Sepal.Length", "Sepal.Width",
"Petal.Length")
lm(Sepal.Length ~., iris[cn])
On 5/17/07, Chris Elsaesser <chris.elsaesser at spadac.com>
wrote:> New to R; please excuse me if this is a dumb question. I tried to RTFM;
> didn't help.
>
> I want to do a series of regressions over the columns in a data.frame,
> systematically varying the response variable and the the terms; and not
> necessarily including all the non-response columns. In my case, the
> columns are time series. I don't know if that makes a difference; it
> does mean I have to call lag() to offset non-response terms. I can not
> assume a specific number of columns in the data.frame; might be 3, might
> be 20.
>
> My central problem is that the formula given to lm() is different each
> time. For example, say a data.frame had columns with the following
> headings: height, weight, BP (blood pressure), and Cals (calorie intake
> per time frame). In that case, I'd need something like the following:
>
> lm(height ~ weight + BP + Cals)
> lm(height ~ weight + BP)
> lm(height ~ weight + Cals)
> lm(height ~ BP + Cals)
> lm(weight ~ height + BP)
> lm(weight ~ height + Cals)
> etc.
>
> In general, I'll have to read the header to get the argument labels.
>
> Do I have to write several functions, each taking a different number of
> arguments? I'd like to construct a string or list representing the
> varialbes in the formula and apply lm(), so to say [I'm mainly a Lisp
> programmer where that part would be very simple. Anyone have a Lisp API
> for R? :-}]
>
> Thanks,
> chris
>
> Chris Elsaesser, PhD
> Principal Scientist, Machine Learning
> SPADAC Inc.
> 7921 Jones Branch Dr. Suite 600
> McLean, VA 22102
>
> 703.371.7301 (m)
> 703.637.9421 (o)
>
> ______________________________________________
> 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.
>
> tmp <- data.frame(matrix(rnorm(40),10,4, dimnames=list(NULL, c("Y","A","B","C")))) > tmp > tmp.form <- paste(names(tmp)[1], paste(names(tmp)[-1], collapse=" + "), sep=" ~ ") > tmp.form > lm(tmp.form, tmp)The R language is powerful enough to most of the lisp-like things you may want to do. Rich
One way to do it is by giving a data frame with the right variables to
lm() as the first argument each time. If lm() is given a data frame as
the first argument, it will treat the first variable as the LHS and the
rest as the RHS of the formula.
As examples, you can do:
lm(myData[c("height", "weight", "BP",
"Cals")])
(The drawback to this is that the "formula" in the fitted model object
looks a bit strange...)
Andy
From: Chris Elsaesser>
> New to R; please excuse me if this is a dumb question. I
> tried to RTFM;
> didn't help.
>
> I want to do a series of regressions over the columns in a data.frame,
> systematically varying the response variable and the the
> terms; and not
> necessarily including all the non-response columns. In my case, the
> columns are time series. I don't know if that makes a difference; it
> does mean I have to call lag() to offset non-response terms. I can not
> assume a specific number of columns in the data.frame; might
> be 3, might
> be 20.
>
> My central problem is that the formula given to lm() is different each
> time. For example, say a data.frame had columns with the following
> headings: height, weight, BP (blood pressure), and Cals
> (calorie intake
> per time frame). In that case, I'd need something like the following:
>
> lm(height ~ weight + BP + Cals)
> lm(height ~ weight + BP)
> lm(height ~ weight + Cals)
> lm(height ~ BP + Cals)
> lm(weight ~ height + BP)
> lm(weight ~ height + Cals)
> etc.
>
> In general, I'll have to read the header to get the argument labels.
>
> Do I have to write several functions, each taking a different
> number of
> arguments? I'd like to construct a string or list representing the
> varialbes in the formula and apply lm(), so to say [I'm mainly a Lisp
> programmer where that part would be very simple. Anyone have
> a Lisp API
> for R? :-}]
>
> Thanks,
> chris
>
> Chris Elsaesser, PhD
> Principal Scientist, Machine Learning
> SPADAC Inc.
> 7921 Jones Branch Dr. Suite 600
> McLean, VA 22102
>
> 703.371.7301 (m)
> 703.637.9421 (o)
>
> ______________________________________________
> 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.
>
>
>
------------------------------------------------------------------------------
Notice: This e-mail message, together with any attachments,...{{dropped}}
I was solving similar problem some time ago.
Here is my script.
I had a data frame, containing a response and several other variables, which
were assumed predictors.
I was trying to choose the best linear approximation.
This approach now seems to me useless, please, don't blame me for that.
However, the script might be useful to you.
<code>
library(forward)
# dfr is a data.frame, that contains everything.
# The response variable is named med5x
# The following lines construct linear models for all possibe formulas
# of the form
# med5x~T+a+height
# med5x~a+height+RH
# T, a, RH, etc are the names of possible predictors
inputs<-names(dfr)[c(10:30,1)] # dfr was a very large data frame,
containing lot of variables.
# here we have chosen only a subset of them.
for(nc in 11:length(inputs)){ # the linear models were assumed to have at
least 11 terms
# now we are generating character vectors containing formulas.
formulas<-paste("med5x",sep="~",
fwd.combn(inputs,nc,fun=function(x){paste(x,collapse="+")}))
# and then, are trying to fit every
for(f in formulas){
lms<-lm(eval(parse(text=f)),data=dfr)
cat(file="linear_models.txt",f,sum(residuals(lms)^2),"\n",sep="\t",append=TRUE)
}
}
</code>
Hmm, looking back, I see that this is rather inefficient script.
For example, the inner cycle can easily be replaced with the apply function.
Chris Elsaesser wrote:>
> New to R; please excuse me if this is a dumb question. I tried to RTFM;
> didn't help.
>
> I want to do a series of regressions over the columns in a data.frame,
> systematically varying the response variable and the the terms; and not
> necessarily including all the non-response columns. In my case, the
> columns are time series. I don't know if that makes a difference; it
> does mean I have to call lag() to offset non-response terms. I can not
> assume a specific number of columns in the data.frame; might be 3, might
> be 20.
>
> My central problem is that the formula given to lm() is different each
> time. For example, say a data.frame had columns with the following
> headings: height, weight, BP (blood pressure), and Cals (calorie intake
> per time frame). In that case, I'd need something like the following:
>
> lm(height ~ weight + BP + Cals)
> lm(height ~ weight + BP)
> lm(height ~ weight + Cals)
> lm(height ~ BP + Cals)
> lm(weight ~ height + BP)
> lm(weight ~ height + Cals)
> etc.
>
> In general, I'll have to read the header to get the argument labels.
>
> Do I have to write several functions, each taking a different number of
> arguments? I'd like to construct a string or list representing the
> varialbes in the formula and apply lm(), so to say [I'm mainly a Lisp
> programmer where that part would be very simple. Anyone have a Lisp API
> for R? :-}]
>
>
--
View this message in context:
http://www.nabble.com/using-lm%28%29-with-variable-formula-tf3772540.html#a10716815
Sent from the R help mailing list archive at Nabble.com.