On 21/03/2009, at 3:19 AM, Ravi Varadhan wrote:
<snip>
I also tried a number of other things including changing the
"family", and parameters in
"loess.control", but to no avail. I looked at the Fortran codes
from both loess and gam.
They are daunting, to say the least. They are dense, and there are
absolutely no comments
whatsoever. But one thing is clear - they are using different
Fortran codes.
Thanks for doing this digging. It would on this basis be not
unreasonable
to expect there to be numerical differences in the result.
So, the best bet might be to get Trevor Hastie or Bill Cleveland to
help you out.
But, before that: why is this an issue, Rolf? Is it important
that these two results
be identical?
It just makes me nervous when two procedures which I believe to be
doing the
same thing give answers which are not identical. Such a phenomenon
makes me
wonder if there is something which I am not understanding, and if
thereby my
lack of understanding might lead to my making serious errors.
A lack of understanding is an all-to-frequent occurrence in my
circumstances
(i.e. being thick as two short planks) and I therefore need to be
constantly
on my guard against such a lack.
In the current setting I think I will desist from pursuing the issue
any further
and just assume that different coding accounts for the different
results. And
take a valium.
cheers,
Rolf Turner
######################################################################
Attention:\ This e-mail message is privileged and confid...{{dropped:9}}
______________________________________________
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.