Message-ID: <51C40C44.9080704@bath.ac.uk>
Date: 2013-06-21T08:18:12Z
From: Simon Wood
Subject: mgcv summary.gam
In-Reply-To: <1177.10.254.253.3.1371798164.squirrel@sqmail.gn.apc.org>
Hi Greg
On 21/06/13 08:02, Greg Dropkin wrote:
> hi
>
> I'm trying to understand (a little) the code behind summary.gam, and have
> the Biometrika article referred to in the Help. But am stuck early on.
>
> In the code, starting at line 167:
>
> if (est.disp)
> rdf <- residual.df
> else rdf <- -1
> res <- testStat(p, Xt, V, df[i], type = p.type,
> res.df = rdf)
>
> 1) shouldn't there be some brackets in the if else statement?
- I don't think so. rdf is either the residual degrees of freedom used
to compute the scale parameter estimate or a signal of fixed scale
parameter, to be passed to testStat.
> 2) what is testStat?
mgcv:::testStat
- probably worth looking at the original source code which includes some
comments. It does the actual computation.
best,
Simon
> thanks
>
> Greg
>
>
>
>
--
Simon Wood, Mathematical Science, University of Bath BA2 7AY UK
+44 (0)1225 386603 http://people.bath.ac.uk/sw283 On
--
Simon Wood, Mathematical Science, University of Bath BA2 7AY UK
+44 (0)1225 386603 http://people.bath.ac.uk/sw283