Re: [Boost-users] [statistics][fusion][mpl][accumulator] chi-square test with boost

Subject: Re: [Boost-users] [statistics][fusion][mpl][accumulator] chi-square test with boost
From: Eric Niebler (eric_at_[hidden])
Date: 2010-05-30 20:16:59

Hi Erwann,

On 5/30/2010 12:47 PM, er wrote:
>>
>> In your case, Boost.Accumulators would be useful to update a
>> contingency table as observations are accumulated. The actual
>> statistics would be in 
> 
> Please find such an accumulator here:
> http://svn.boost.org/svn/boost/sandbox/statistics/non_parametric/boost/statistics/detail/non_parametric/contingency_table/

>From just a glance it looks like it could be useful, but it's too much
for me to digest. It will certainly needs docs and tests. Perhaps that
would help.

> An example side by side with R code
> https://svn.boost.org/svn/boost/sandbox/statistics/non_parametric/libs/statistics/detail/non_parametric/test/contingency_table2.cpp
> 
> Warning : tested under OSX only, for now.
> 
> This has necessitated extensions:
> http://svn.boost.org/svn/boost/sandbox/statistics/support/boost/accumulators/

I don't see what changed to weighted_count.hpp besides your changing the
comment at the top to "factor.hpp", deleting my copyright (!!!), and
commenting out the weighted_count_of_variates features and extractor.
And what is the purpose of unpack_depends_on.hpp?

> http://svn.boost.org/svn/boost/sandbox/statistics/support/boost/fusion/
> http://svn.boost.org/svn/boost/sandbox/statistics/support/boost/mpl/
> 
> May the authors of the relevant library please notify me if they think
> feature requests are warranted?
> 
> Thanks.

-- 
Eric Niebler
BoostPro Computing
http://www.boostpro.com