# Re: [R] Prediction when using orthogonal polynomials in regression

From: Achim Zeileis <Achim.Zeileis_at_wu-wien.ac.at>
Date: Fri 27 Jan 2006 - 22:31:41 EST

On Thu, 26 Jan 2006 22:10:23 +0530 Ajay Narottam Shah wrote:

> Folks,
>
> I'm doing fine with using orthogonal polynomials in a regression
> context:
>
> # We will deal with noisy data from the d.g.p. y = sin(x) + e
> x <- seq(0, 3.141592654, length.out=20)
> y <- sin(x) + 0.1*rnorm(10)

> d <- lm(y ~ poly(x, 4))
> plot(x, y, type="l"); lines(x, d\$fitted.values, col="blue")

fitted(d) is usually the preferred way of accessing the fitted values (although equivalent in this particular case).

> great! all.equal(as.numeric(d\$coefficients + m %*% d\$coefficients
> [2:5]), as.numeric(d\$fitted.values))
>
> What I would like to do now is to apply the estimated model to do
> prediction for a new set of x points e.g.
> xnew <- seq(0,5,.5)
>
> We know that the predicted values should be roughly sin(xnew). What I
> don't know is: how do I use the object `d' to make predictions for
> xnew?

Use predict:
predict(d, data.frame(x = xnew))
which is pretty evocative.

Best,
Z

> --
> Ajay Shah
> http://www.mayin.org/ajayshah
> ajayshah@mayin.org
> http://ajayshahblog.blogspot.com <*(:-? - wizard who doesn't know the
>
> ______________________________________________
> R-help@stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help