isemaj
Reputation: 587
Degrees of freedom in the smoothing spline fit in a GAM formula
If the smoothing spline is simply a natural cubic spline with knots at every unique value of x_i. Then why does gam::s() in R need a degree of freedom?
Upvotes: 0
Views: 523
Answers (1)
Zheyuan Li
Reputation: 73265
Because it is smoothing not interpolation. The degree of freedom tells how much complexity you want in the fitted spline.
Upvotes: 2
Related Questions
- Fitting a smooth spline (GAM function) in R: Error in number of knots required to fit spline -- knot requirements increasing
- Optimizing degrees of freedom in spline regression
- B Spline confusion
- smooth.spline(): fitted model does not match user-specified degree of freedom
- Different python functions to fit cubic splines, finding coefficients
- R smooth.spline(): smoothing spline is not smooth but overfitting my data
- natural cubic spline regression with R
- fit a smoothing spline using gam function
- Error in gam function of smoothing spline regression
- specifying degrees of freedom for b-spline fit using bs function in splines package