[R] mgcv, gam

[ripley@stats.ox.ac.uk]

On Fri, 31 Jan 2003, Simon Wood wrote:

> > I have some problems with gam in mgcv. Firts a detail: it would
> > be nice igf gam would accept an na.action argument, but that not the 
> > main point.
> - I find it hard to think of a sensible action except dropping the
> associated data, but if you've a concrete suggestion I'm happy to add it
> to the "to do" list.

Simon: there's a standard interface here: you call na.action and it
decides what to do.  Basically you need to set up a model frame as e.g.
lm does, and then pass that as `data' to gam.setup.  It would eliminate
some other inconsistencies (e.g. not using environment(formula)) too.

I hope that's clear enough, but please feel to talk to me privately.

Brian

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595