Bug #97925 for PDL-Stats: Mixed ANOVA with 3-level between-subjects variable -- uninvertible matrix

Sun Aug 10 20:38:22 2014 MAGGIEXYZ [...] cpan.org - Ticket created

CC:	erich.greene [...] yale.edu
Subject:	Mixed ANOVA with 3-level between-subjects variable -- uninvertible matrix

From Erich Greene: Following up on the within factor masquerading as a between, I took a look at the original mixed model test (t_anova_rptd_mixed) and found the problem there as well: $a and $b are both within-subject factors. So working with the intended 3-level within * 2-level between model from that test (changing the original subject vector from 0..3x6 to 0..7x3 makes $b between-subjects), I again found a dependence on the order of the data within the piddles. (In principle, we should get the same results no matter how we permute the input vectors, as long as we apply the same permutation to all of them.) While it runs with both orderings, the output is only correct when the data is grouped by subject (my frontends all do this, which is why I'd never noticed a problem before); when the data is grouped by within var level, the error sums of squares are wrong (though the degrees of freedom are correct, unlike in the 3-level between case).

Sun Aug 10 20:39:29 2014 MAGGIEXYZ [...] cpan.org - Status changed from 'new' to 'open'

Sun Aug 10 20:52:24 2014 MAGGIEXYZ [...] cpan.org - Correspondence added

The problem is in the _add_errors() subroutine for anova_rptd(). Wish I had left more comments there!

Mon May 30 15:44:34 2016 ETJ [...] cpan.org - Correspondence added

Maggie / Erich, Do you have a short snippet of code that will demonstrate this?

Tue May 31 13:44:22 2016 erich.greene [...] yale.edu - Correspondence added

Subject:	RE: [rt.cpan.org #97925] Mixed ANOVA with 3-level between-subjects variable -- uninvertible matrix
Date:	Tue, 31 May 2016 17:43:54 +0000
To:	"bug-PDL-Stats [...] rt.cpan.org" <bug-PDL-Stats [...] rt.cpan.org>, "MAGGIEXYZ [...] cpan.org" <MAGGIEXYZ [...] cpan.org>
From:	"Greene, Erich" <erich.greene [...] yale.edu>

Hi Ed, I submitted a new test suite once upon a time, but I no longer have access to my files from back then (changed jobs) and it looks like the pdl mailing list archive doesn't go back that far. Maybe Maggie still has that code? -- Erich Erich J. Greene, PhD Biostatistician Yale Center for Analytical Sciences 300 George St, Suite 555 New Haven, CT 06511 203-737-3636 Show quoted text

-----Original Message----- From: Ed J via RT [mailto:bug-PDL-Stats@rt.cpan.org] Sent: Monday, May 30, 2016 3:45 PM To: MAGGIEXYZ@cpan.org Cc: Greene, Erich <erich.greene@yale.edu> Subject: [rt.cpan.org #97925] Mixed ANOVA with 3-level between-subjects variable -- uninvertible matrix <URL: https://rt.cpan.org/Ticket/Display.html?id=97925 > Maggie / Erich, Do you have a short snippet of code that will demonstrate this?

Tue May 31 16:28:43 2016 zaki.mughal [...] gmail.com - Correspondence added

CC:	;
Subject:	Re: [rt.cpan.org #97925] Mixed ANOVA with 3-level between-subjects variable -- uninvertible matrix
Date:	Tue, 31 May 2016 15:28:25 -0500
To:	"erich.greene [...] yale.edu via RT" <bug-PDL-Stats [...] rt.cpan.org>
From:	Zakariyya Mughal <zaki.mughal [...] gmail.com>

On 2016-05-31 at 13:44:23 -0400, erich.greene@yale.edu via RT wrote: Show quoted text

> Queue: PDL-Stats > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=97925 > > > Hi Ed, > > I submitted a new test suite once upon a time, but I no longer have > access to my files from back then (changed jobs) and it looks like the > pdl mailing list archive doesn't go back that far. Maybe Maggie still > has that code?

Hi, Would it happen to the be the same test posted as an attachment in the thread here <http://comments.gmane.org/gmane.comp.lang.perl.pdl.general/8279>? Cheers, - Zaki Mughal Show quoted text

> > -- Erich > > Erich J. Greene, PhD > Biostatistician > Yale Center for Analytical Sciences > 300 George St, Suite 555 > New Haven, CT 06511 > 203-737-3636 > > -----Original Message----- > From: Ed J via RT [mailto:bug-PDL-Stats@rt.cpan.org] > Sent: Monday, May 30, 2016 3:45 PM > To: MAGGIEXYZ@cpan.org > Cc: Greene, Erich <erich.greene@yale.edu> > Subject: [rt.cpan.org #97925] Mixed ANOVA with 3-level between-subjects variable -- uninvertible matrix > > <URL: https://rt.cpan.org/Ticket/Display.html?id=97925 > > > Maggie / Erich, > > Do you have a short snippet of code that will demonstrate this? >

Tue May 31 16:36:35 2016 erich.greene [...] yale.edu - Correspondence added

Subject:	RE: [rt.cpan.org #97925] Mixed ANOVA with 3-level between-subjects variable -- uninvertible matrix
Date:	Tue, 31 May 2016 20:36:19 +0000
To:	"bug-PDL-Stats [...] rt.cpan.org" <bug-PDL-Stats [...] rt.cpan.org>, "MAGGIEXYZ [...] cpan.org" <MAGGIEXYZ [...] cpan.org>
From:	"Greene, Erich" <erich.greene [...] yale.edu>

Hi Zaki, Yes, I think REVISED_mixed_anova_test.pl was my last stab at test code for it. Thanks, -- Erich Show quoted text

-----Original Message----- From: Zakariyya Mughal via RT [mailto:bug-PDL-Stats@rt.cpan.org] Sent: Tuesday, May 31, 2016 4:29 PM To: MAGGIEXYZ@cpan.org Cc: Greene, Erich <erich.greene@yale.edu> Subject: Re: [rt.cpan.org #97925] Mixed ANOVA with 3-level between-subjects variable -- uninvertible matrix <URL: https://rt.cpan.org/Ticket/Display.html?id=97925 > On 2016-05-31 at 13:44:23 -0400, erich.greene@yale.edu via RT wrote:

> Queue: PDL-Stats > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=97925 > > > Hi Ed, > > I submitted a new test suite once upon a time, but I no longer have > access to my files from back then (changed jobs) and it looks like the > pdl mailing list archive doesn't go back that far. Maybe Maggie still > has that code?

Hi, Would it happen to the be the same test posted as an attachment in the thread here <https://urldefense.proofpoint.com/v2/url?u=http-3A__comments.gmane.org_gmane.comp.lang.perl.pdl.general_8279&d=AwIDaQ&c=-dg2m7zWuuDZ0MUcV7Sdqw&r=iAcL2PTMvR7scY1NHx7d1JkZWWFqroT_9N2n5Y1RsBs&m=ezT4IIYmCAlGkoXS2gXTPwqoKw8psCPaMQDkF98dp3c&s=GVn-PMKIv9nZ2nAcuiGJ2yZjVENKlC1vI0_04_gsiEY&e= >? Cheers, - Zaki Mughal

> > -- Erich > > Erich J. Greene, PhD > Biostatistician > Yale Center for Analytical Sciences > 300 George St, Suite 555 > New Haven, CT 06511 > 203-737-3636 > > -----Original Message----- > From: Ed J via RT [mailto:bug-PDL-Stats@rt.cpan.org] > Sent: Monday, May 30, 2016 3:45 PM > To: MAGGIEXYZ@cpan.org > Cc: Greene, Erich <erich.greene@yale.edu> > Subject: [rt.cpan.org #97925] Mixed ANOVA with 3-level > between-subjects variable -- uninvertible matrix > > <URL: https://rt.cpan.org/Ticket/Display.html?id=97925 > > > Maggie / Erich, > > Do you have a short snippet of code that will demonstrate this? >

Tue May 31 21:34:18 2016 maggiexyz [...] gmail.com - Correspondence added

CC:	MAGGIEXYZ [...] cpan.org
Subject:	RE: [rt.cpan.org #97925] Mixed ANOVA with 3-level between-subjects variable -- uninvertible matrix
Date:	Tue, 31 May 2016 21:33:49 -0400
To:	bug-PDL-Stats [...] rt.cpan.org
From:	Maggie X <maggiexyz [...] gmail.com>

Yes I have the test script. Attached. Honestly I was a little taken aback when I looked at my code for this thing :P I meant to just rewrite this whole function but never had the opportunity to get into the state required. Maybe I will have more time in a few more months... Thanks for checking in! Maggie On May 31, 2016 1:44 PM, "erich.greene@yale.edu via RT" < bug-PDL-Stats@rt.cpan.org> wrote: Show quoted text

> <URL: https://rt.cpan.org/Ticket/Display.html?id=97925 > > > Hi Ed, > > I submitted a new test suite once upon a time, but I no longer have access > to my files from back then (changed jobs) and it looks like the pdl mailing > list archive doesn't go back that far. Maybe Maggie still has that code? > > -- Erich > > Erich J. Greene, PhD > Biostatistician > Yale Center for Analytical Sciences > 300 George St, Suite 555 > New Haven, CT 06511 > 203-737-3636 > > -----Original Message----- > From: Ed J via RT [mailto:bug-PDL-Stats@rt.cpan.org] > Sent: Monday, May 30, 2016 3:45 PM > To: MAGGIEXYZ@cpan.org > Cc: Greene, Erich <erich.greene@yale.edu> > Subject: [rt.cpan.org #97925] Mixed ANOVA with 3-level between-subjects > variable -- uninvertible matrix > > <URL: https://rt.cpan.org/Ticket/Display.html?id=97925 > > > Maggie / Erich, > > Do you have a short snippet of code that will demonstrate this? > >

Message body is not shown because sender requested not to inline it.