[R] Stepwise

[arun]

Hi,
May be this helps:
set.seed(232)
mat1<-matrix(sample(1:100,80,replace=TRUE),ncol=8) #with 8 columns
dat1<-data.frame(mat1)
 names(dat1)[1]<-"Y"
 form1<-as.formula(paste("Y~",paste(names(dat1)[2:8],collapse="+"))) #should change "8" to "400"

fit.final<-lm(form1,data=dat1)

fit.final
#Call:
#lm(formula = form1, data = dat1)
#
#Coefficients:
#(Intercept)           X2           X3           X4           X5           X6  
# 143.763081    -0.009030    -1.844810     1.436647    -0.005407    -0.050101  
#         X7           X8  
#   0.563285    -1.901918  
A.K.






________________________________
From: farnoosh sheikhi <farnoosh_81 at yahoo.com>
To: arun <smartpink111 at yahoo.com> 
Sent: Thursday, November 15, 2012 6:14 PM
Subject: Stepwise


Hi there,

I want to apply the following code fo my data with 400 predictors.
I was wondering if there is an alternative way instead of typing 400 predictors for the following code.
I really appreciate your help.

fit0<-lm(Y~1, data= mydata)
fit.final<- lm(Y~X1+X2+X3+.....+X400, data=mydata) ???
step(fit0, scope=list(lower=fit0, upper=fit.final), data=mydata, direction="forward")
step(fit.final, scope=list(lower=fit.final, upper=fit0), data=mydata, direction="backward")

Best,Farnoosh Sheikhi