R help - Mar 2007 - simple multivariate linear model plot

Dear R-users,

I am trying to get a simple plot for a linear model. The following function does
the job
of getting all the plots of the lm function but the "main" plot.
 
multivarplot <-
  function(...){
    x1 <- rnorm(10); x2 <- rnorm(10); x3 <- rnorm(10)
    y  <- rnorm(10)
    multivarfit <- lm(y ~ x1 + x2 + x3)
    par(mfrow = c(3,2))
    plot.lm(multivarfit)
  }

where the "main" plot is graph of the variable (fitted : x1 + x2 + x3
) against y.
This is similar to the graph in Phil Spector's book (intro to the S and
Splus) p. 227.
It seems to be a very simple task but I cannot figure it out.
Thank you for any hint.
I use R 2.4.0 windows xp.

AA.

	[[alternative HTML version deleted]]

I think I found the answer to my own question.
One can use 
fittedvalue <- fitted(multivarfit)
plot(fittedvalue, y) in the below function.
Thanks to Verzani's simple R manual for leading me to this.

  ----- Original Message ----- 
  From: AA 
  To: R-Help 
  Sent: Sunday, March 18, 2007 2:42 PM
  Subject: simple multivariate linear model plot


  Dear R-users,

  I am trying to get a simple plot for a linear model. The following function
does the job
  of getting all the plots of the lm function but the "main" plot.

  multivarplot <-
    function(...){
      x1 <- rnorm(10); x2 <- rnorm(10); x3 <- rnorm(10)
      y  <- rnorm(10)
      multivarfit <- lm(y ~ x1 + x2 + x3)
      par(mfrow = c(3,2))
      plot.lm(multivarfit)
    }

  where the "main" plot is graph of the variable (fitted : x1 + x2 +
x3 ) against y.
  This is similar to the graph in Phil Spector's book (intro to the S and
Splus) p. 227.
  It seems to be a very simple task but I cannot figure it out.
  Thank you for any hint.
  I use R 2.4.0 windows xp.

  AA.

	[[alternative HTML version deleted]]