gam.parser {mgcv}R Documentation

Generalized Additive Models fitting using penalized regression splines and GCV

Description

This is an internal function of package mgcv. It is a service routine for gam which splits a GAM formula into two parts: one representing the smooth part of the model and the other representing the regular parametric part of the model (including any intercept term). The processing of the smooth terms in the model also returns the number of knots for each smooth term, and whether or not the smooth term is to be treated as a pure regression spline (i.e. one without a penalty). For further information on usuage see code for gam.

WARNING

The code does not check for rank defficiency of the model matrix that may result from including the same covariates in the smooth and parametric parts of the model - it will likely just fail instead!

Author(s)

Simon N. Wood snw@st-and.ac.uk

References

Gu and Wahba (1991) Minimizing GCV/GML scores with multiple smoothing parameters via the Newton method. SIAM J. Sci. Statist. Comput. 12:383-398

Wood (2000) Modelling and Smoothing Parameter Estimation with Multiple Quadratic Penalties. JRSSB 62(2):413-428

http://www.ruwpa.st-and.ac.uk/simon.html

See Also

gam mgcv