Bingzhang Chen
2011-Aug-06 00:59 UTC
[R] How to estimate confidential intervals for the derivatives of cubic smoothing spline
Dear all, I want to use smooth.spline to construct a cubic smoothing spline and its first derivative to my data. However, the predict.smooth.spline does not seem to provide a SE for both the fitted values and their derivatives. How should I calculate it? Thank you very much, Bingzhang
Simon Wood
2011-Aug-06 05:38 UTC
[R] How to estimate confidential intervals for the derivatives of cubic smoothing spline
If you can put up with using gam in package mgcv to fit your splines then you can get SEs on the spline derivatives as shown in the final example in the help file for plot.gam. (gam uses reduced rank smoothing splines, so you may need to up the `k' parameter for the smooth terms to ensure that the rank is not too small for your data). On 08/06/2011 01:59 AM, Bingzhang Chen wrote:> Dear all, > > I want to use smooth.spline to construct a cubic smoothing spline and its first > derivative to my data. However, the predict.smooth.spline does not seem to > provide a SE for both the fitted values and their derivatives. How should I > calculate it? > > > Thank you very much, > Bingzhang > > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >