Thanks for the help. I tried out the one promising lead, curfit.free.knot,
and it doesn't work for linear or quadratic splines. The documentation says
it should, but when I specify a linear spline, it returns a cubic.
On 5/1/08, Spencer Graves <[EMAIL PROTECTED]> wrote:
>
> RSiteSearch('fr
Suppose I have two variables, x and y. For a fixed number of knots, I want
to create a spline transformation of x such that a loss function is
minimized. Presumably, this loss function would be least squares, i.e. sum
(f(x)-y)^2. The spline transformations would be linear, quadratic or
cubic. I
Great suggestions. I tested the code on an example and the run time was
reduced from 1 min 12 sec to 3 sec. Also, I like the suggestion to look at
the quantiles. I will see what insight it provides in terms of detecting
masked interactions.
I have a couple questions about your code.
First, why
The code I wrote does this using predict() which is useful for modeling
approaches like GAMs.
Mike
On Wed, Apr 23, 2008 at 8:47 PM, hadley wickham <[EMAIL PROTECTED]> wrote:
> On Wed, Apr 23, 2008 at 7:31 PM, Mike Dugas <[EMAIL PROTECTED]> wrote:
> > Thanks for the help. Th
t 4:23 PM, Mike Dugas <[EMAIL PROTECTED]> wrote:
> > The answer to my post is yes (which I just figured out).
> >
>
> Switching from for to apply isn't going to speed up your code. If you
> carefully read the source code of apply, you'll see the guts
, 1112, byrow=T)
a[2,] <- apply(b,2,FUN=function(x)
{mean(predict(lm1,cbind(m[,-match("x1",names(m))],x1=x))) })
plot(a[1,],a[2,],xlab="x1",ylab="Response",type="l",main="Partial Dependence
Plot")
Mike Dugas
[[alternative HTML version
Hey all,
The code below creates a partial dependence plot for the variable x1 in the
linear model y ~ x1 + x1^2 + x2.
I have noticed that the for loop in the code takes a long time to run if the
size of the data is increased. Is there a way to change the for loop into
an apply statement? The tr
7 matches
Mail list logo
