Mailing-List: contact dev-help@commons.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Commons Developers List" <dev@commons.apache.org>
Received-SPF: pass (nike.apache.org: domain of phil.steitz@gmail.com
 designates 209.85.160.43 as permitted sender)
Message-ID: <5096C853.1050309@gmail.com>
Date: Sun, 04 Nov 2012 11:56:03 -0800
From: Phil Steitz <phil.steitz@gmail.com>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7;
 rv:16.0) Gecko/20121026 Thunderbird/16.0.2
MIME-Version: 1.0
To: Commons Developers List <dev@commons.apache.org>
Subject: Re: [Math] MATH-878: Feature request with patch
References: <20121020105843.GC31848@dusk.harfang.homelinux.org>
 <5084E2A3.8070304@gmail.com>
 <CAJwFCa21HV2kbstF-VkE7ZjQrUyDoQ4+SZT-MBFnt0AYh52e=g@mail.gmail.com>
 <5084E8E5.5000500@gmail.com>
 <CAJwFCa3QxQ6VrEp55v2_m+0LdpmPy6AxRWkxDsmgLyzh==bnog@mail.gmail.com>
In-Reply-To: 
 <CAJwFCa3QxQ6VrEp55v2_m+0LdpmPy6AxRWkxDsmgLyzh==bnog@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit

On 10/22/12 8:15 AM, Ted Dunning wrote:
> On Sun, Oct 21, 2012 at 11:34 PM, Phil Steitz <phil.steitz@gmail.com> wrote:
>
>> On 10/21/12 11:25 PM, Ted Dunning wrote:
>>> What kind of check did you want?
>>>
>>> I checked the code by eye and supplied several test cases.  You might say
>>> that I am versed in statistics since I am the author of the major paper
>> on
>>> this test as applied to computational linguistics.
>> I was going to mention that :)
>>
>> Have you carefully reviewed the code?
>>
> I have pretty high confidence in it.  The algorithm is the simplest that I
> know (increases likelihood of correctness) and he seems to have
> incorporated my test cases.
>
>
>> Thanks in advance if you have time.  I will look at it as well soon
>> and take a stab at moving some of the reference material into the
>> javadoc.  Thanks in any case for helping move this along.
>>
> Thanks for that.
>
Sorry it took me so long to get this committed.  It took me longer
than I expected to get myself educated.  I got a lot out of [1] and
thank you for writing it, Ted.  The bigram example there very nicely
illustrates how ChiSquare stats can be misleading.  You mention at
the end that Fisher's exact test might also be used in these
situations.  I am curious about the following:

0) Did you or anyone else ever analyze the bigram data in the paper
using Fisher's test stats?
1) Is the bigram data from [1] available anywhere?
1) Do you think a direct implementation of Fisher's test for 2x2
designs and a monte carlo impl for r x c would be useful?  I have
this in C from years ago and could translate it fairly easily.

Phil

[1] http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.14.5962

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org