R help - Jan 2012 - Linear regression using matrices

I'm new to R and I'm not a Statistician I'm an Accountant, but
I'm finding it
an excellent tool for the business analysis work I do.

I need to run LM() where both response and predictor are held in matrices.
The model follows the form:-
regression1 = matrix1.col1 <-> matrix2.col1
regression2 = matrix1.col2<->matrix2.col2
and so on......

I have no problem with the multivariate response but I can't get the
function to recognise a tabular predictor.

Is this possible without using a loop ?

--
View this message in context:
http://r.789695.n4.nabble.com/Linear-regression-using-matrices-tp4260945p4260945.html
Sent from the R help mailing list archive at Nabble.com.

I've done a lot of research on this very topic and found a few solutions.
But
all the ways I've discovered involve loops.

Applying it to what you want, the best way I've found is to do (stolen from
an experienced R user, of course):

y<-array(rnorm(100),dim=c(10,10))
x<-array(rnorm(100),dim=c(10,10))

regg<-list()
for(i in 1:ncol(y))
 {
 regg[[i]]<-lm(y[,i]~x[,i])
 }

Now the reason I've stuck onto the list() method is because the output is
still in lm() format. So you can write neat functions to extract every
statistic you want off summary(regg[[N]]). 

-----
----

Isaac
Research Assistant
Quantitative Finance Faculty, UTS
--
View this message in context:
http://r.789695.n4.nabble.com/Linear-regression-using-matrices-tp4260945p4260988.html
Sent from the R help mailing list archive at Nabble.com.

OK thanks.

In my case I think it might be possible to work around this by reshaping my
data and then using lmlist() to run separate regressions for each data
group. lmlist() is new to me but it looks like it will do the job.


--
View this message in context:
http://r.789695.n4.nabble.com/Linear-regression-using-matrices-tp4260945p4261368.html
Sent from the R help mailing list archive at Nabble.com.