[R] Robust GAM that covers Gaussian Distributions

[Simon Wood]

The 'scat' (scaled t) family in mgcv does allow for much heavier tails 
than Gaussian, but it may not be robust enough for you.

On 25/02/15 15:27, Ilaria Prosdocimi wrote:
> Ilaria Prosdocimi <ilapro <at> ceh.ac.uk> writes:
>
>>
>> Christos Giannoulis <cgiannoul <at> gmail.com> writes:
>>
>>>
>>> Dear All,
>>>
>>> I was looking the r-archives and crantastic...for a package that has
> a
>>> robust approach to generalized additive models. I found two packages
>>> "robustgam" and "rgam" but their implemented functions
>>> cover only binomial and poisson distributions (pls correct me if I
> am
>>> wrong).
>>>
>>> I would greatly appreciate if anyone could share with us other
> packages or
>>> robust approaches of general additive modeling that might have a
> better
>>> performance with small data sets (n = 50 -100 records).
>>>
>>> Thank you very much all for reading this message. I am hoping and
> looking
>>> forward to receiving your reply.
>>>
>>> Sincerely,
>>>
>>> Christos Giannoulis
>>>
>>> 	[[alternative HTML version deleted]]
>>>
>>>
>>
>> Indeed, it seems that both the libraries do not allow normal data. I
> think
>> you could try to use the code in this page
>> (http://www.stat.ubc.ca/~matias/penalised/) for S-estimation.
>> If you want the code for the Croux et al paper
>> (http://onlinelibrary.wiley.com/doi/10.1111/j.1541-
> 0420.2011.01630.x/full)
>> just email me (ilapro + ceh.ac.uk) - it seems to be not available
> online
>> anymore. This also allow normal data.
>>
>> Best
>>
>> Ilaria
>>
>>
>
>
>
> For information - in case somebody finds this post again in the future,
> the original files have now been restored in the KULeuven website at
> this address
> http://wis.kuleuven.be/stat/stat-inferen/codes
>
> I have also packaged the code in a R package which can be accessed on
> GitHub (and installed with devtools::install_github) at
>
> https://github.com/ilapros/DoubleRobGam
>
> Best
>
> Ilaria
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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.
>


-- 
Simon Wood, Mathematical Science, University of Bath BA2 7AY UK
+44 (0)1225 386603               http://people.bath.ac.uk/sw283