Thank you for your answer !
Actually I'm doing an internship about GAM for mid-term french load forecasting
(at "EDF", Electricité De France). I'm working with your book and I was asked
to simulate data on my own, in order to see if gam (from mgcv) gave good
estimations of the functions I had used for simulation, even with explicative
variables like the load of the previous day, or with correlated noise.
We knew that it was possible to get the coefficients and so, at this moment of
our work, we wondered if there was a way to get the actual bases in order to
estimate a "distance" between our original functions and the functions
estimated by gam. In the meantime, I have created a new "instrumental" data
frame with many points from the definition areas of my original functions, and
used predict.gam on it with link="terms".
Amandine.
_________________________________________________________________
Retouchez, classez et partagez vos photos gratuitement avec le logiciel Galerie
de Photos !
[[alternative HTML version deleted]]
______________________________________________
R-help@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.
