calories  <- c(110,110,150,130,120)
fat   <- c(0,1,3,2,1)
Cereal <- data.frame(calories,fat)
rownames(Cereal) <- c("Crispix","Froot Loops","Muesli Raisins","Oatmeal Raisin","Post Nat Raisin")
Cereal
rm(calories,fat)                                  # remove unneeded objects
Cereal.Fit <- lm(Cereal$calories ~ Cereal$fat)    # fit a linear regression model


  # the output from SAS Proc Reg is contained in one or the other of summary() or anova()
summary(Cereal.Fit)
anova(Cereal.Fit)


  # a plot with the raw data, the fitted regression line, and a lowess line
plot(Cereal$fat,Cereal$calories,xlab="Fat",ylab="Calories",main="Regression with cereal data")
abline(Cereal.Fit)
lines(lowess(Cereal$fat,Cereal$calories))

  # this code creates a residual-by-predicted plot and a normal plot of the residuals
par(mfrow=c(2,1))                                 # makes all future plots appear two 
                                                  # per page, vertically
plot(fitted(Cereal.Fit),resid(Cereal.Fit),xlab="Predicted Values",
     ylab="Residuals",main="Residual by Predicted plot")
qqnorm(resid(Cereal.Fit))
qqline(resid(Cereal.Fit))
par(mfrow=c(1,1))                                 # restores one plot per page


  # this code produces a plot like the one in the corresponding SAS code
  # note that the function 'predict' can produce predicted values, 
  #                confidence interval values, and prediction interval values
plot(Cereal$fat,Cereal$calories,xlab="Fat",ylab="Calories",main="Regression with cereal data",
      xlim=c(0,3),ylim=c(75,180))
points(Cereal$fat,predict(Cereal.Fit),pch="r")
points(Cereal$fat,predict(Cereal.Fit, interval="confidence")[,2],pch="c")
points(Cereal$fat,predict(Cereal.Fit, interval="confidence")[,3],pch="c")
points(Cereal$fat,predict(Cereal.Fit, interval="prediction")[,2],pch="p")
points(Cereal$fat,predict(Cereal.Fit, interval="prediction")[,3],pch="p")

  # here we print out the confidence and prediction interval values, compare to the SAS printout
predict(Cereal.Fit, interval="confidence")
predict(Cereal.Fit, interval="prediction")