Plot Regression Surface

Question

I am reading a book by Cohen, Cohen, Aiken and West(2003) \"Applied Multiple Regression Correlation Analysis for the Behavioral Sciences\" and have come across a 3d plot of

Answer 1

Here's how I would do it (adding a bit of color) with packages 'rms' and 'lattice':

require(rms)  # also need to have Hmisc installed
require(lattice)
ddI <- datadist(DF)
options(datadist="ddI")
lininterp <- ols(y ~ x1*x2, data=DF)
 bplot(Predict(lininterp, x1=25:40, x2=45:60), 
       lfun=wireframe,  # bplot passes extra arguments to wireframe
       screen = list(z = -10, x = -50), drape=TRUE)

And the non-interaction model:

 bplot(Predict(lin.no.int, x1=25:40, x2=45:60), lfun=wireframe, col=2:8, drape=TRUE, 
screen = list(z = -10, x = -50),
 main="Estimated regression surface with no interaction")