[R] Modifying a design matrix in mgcv

[David Winsemius]

On Aug 14, 2013, at 12:46 PM, Andrew Crane-Droesch wrote:

> Thanks!  I hadn't touched environment variables before, not knowing what they were.  The G$X matrix indeed seems to get squared, giving appropriately nonsensical results.  I can now go and code a (hopefully) sensible change to it.

Terminology caution: That is not manipulating what most people would call "environment variables", but is rather "assigning an environment to a function". The phrase "environment variables" generally refer to objects or OS settings in which R itself is operating.

You can get more information about "environment variables" with

??"environment variables"

> 
> 
> On 08/14/2013 10:32 PM, David Winsemius wrote:

>> environment(mod.gam) <- environment(gam)
> 

David Winsemius
Alameda, CA, USA