[R] Fitting spline using Pspline
guy33
david.reshef at magd.ox.ac.uk
Mon May 30 00:28:18 CEST 2011
Ravi,
Thanks so much! You're right, smooth.spline does work on larger n.
Although, for some reason it's results are different (slightly less good?,
but I'm not sure). For example, on the simple doppler function below,
sm.spline seems to be closer to the true function than smooth.spline:
x=array(0,1000)
y=array(0,1000)
for (i in 1:1000){
x[i] = i/1000
y[i] = (x[i]*(1-x[i]))^.5 * sin(2*pi*(1.05/(x[i]+.05)))
}
plot(x,y)
fit = sm.spline(x, y, norder=2, cv=FALSE)
lines(fit$x,fit$y)
fit2 = smooth.spline(x, y, cv=FALSE)
lines(fit2$x,fit2$y)
What do you make of that?
-guy33
--
View this message in context: http://r.789695.n4.nabble.com/Fitting-spline-using-Pspline-tp3559202p3559610.html
Sent from the R help mailing list archive at Nabble.com.
More information about the R-help
mailing list