Re: [R] Smoothing Spline Basis Matrix

Thanks, Derek. I used your code, but H still has a lot of columns being 0. I'm not sure why. I'm dealing the data http://www-stat.stanford.edu/~tibs/ElemStatLearn/ with X=age, Y=spnbmd I think X has too many duplicated data??? I'll check the book you recommended in the library tomorrow. Thanks

Re: [R] Smoothing Spline Basis Matrix

Thank you. D Sonderegger. Did you mean use bs <- bs(X, knots = spl$fit$knot) H <- predict(bs, X) Then H should be the matrix for the original data under the smoothing spline basis? However, another problem arises, since I need to use H to estimate the coefficient beta, which involves (H'H)^{-

Re: [R] Smoothing Spline Basis Matrix

Close, but not quite. Your design matrix is: H <- bs(X, knots=spl$fit$knot, intercept=TRUE) So long as you have data in all of the inter-knot regions, you should be ok. A good book to check out about this sort of thing is Ruppert, Wand and Carroll's Semiparametric Regression book. I can't reco

Re: [R] Smoothing Spline Basis Matrix

I believe that smooth.spline fits a cubic B-spline to the data. So you just need to know the knot points (which are returned by smooth.spline as spl$fit$knot) and then use the bs() function in the splines library. pinkdd wrote: > > Anybody knows how to generate the basis matrix for smoothing s

[R] Smoothing Spline Basis Matrix

Anybody knows how to generate the basis matrix for smoothing spline? And how about the smoother matrix? I tried to use the following code, but there exist replicated data in X, and the length of smooth.spline(X, S[,i])$y is smaller than X, and then there is error in the last step. spl <- smoot