[R] gamm(): degrees of freedom of the fit

[Thomas Lumley]

On Wed, 1 Nov 2006, Fabian Scheipl wrote:
>
> Calculating the smoother/projection matrix S: y -> \hat y and then its 
> trace by sum(eigen(S))$values is what I've been doing so far- but I was 
> hoping there might be a more efficient way than doing the spectral 
> decomposition of an NxN-matrix.
>

Well, there are more efficient ways to compute the trace of a matrix than 
spectral decomposition. The trace is just the sum of the diagonal 
elements: sum(diag(S))

 	-thomas