Re: Singularity In Average Information Matrix Using Ainverse

Dear Roman,

Your AI singularity problem is because your model is overspecified with
respect to your data. It is pointing to a different problem to that
which the earlier emails discussed.

I conclude that your HTD fixed effect is completely confounded with
animals.

Although your pedigree has 341000 individuals in it, only 44102
individuals have data (341000 - 296898 effects in ide(anim).

It is odd that anim also has 296898 zero effects, meaning that
there are no individuals in the pedigree without data that are related
to animals with data.

You have 692840 records, and have 640384 residual DF after fixed effects
(i.e. you have fitted 52456 fixed effects). 52424 fixed effects are
associated with HTD. There are substantially more levels of HTD than
fixed individuals with data.

So I conclude animals are confounded with HTD. Hence there is no
information to estimated ide(anim).

Using !AISING will not help you. it just means that the analysis will
articially partition ide(anim) variation put some in HTD and leaving
some in ide(anim).

On Thu, 2009-09-24 at 05:48 -0700, Sergio I. Roman Ponce wrote:
> Dear Dr. Arthur
>
> I am running ASREML 2 but I am having those results. i am warming
> about the " Singularity in Average Information Matrix".
>
> This is the part of the output:
>
> Forming 734460 equations: 36 dense.
> Initial updates will be shrunk by factor 0.032
> Notice: LogL values are reported relative to a base of -720000.00
> Notice: 4 singularities detected in design matrix.
> Notice: 1 singularities appeared in Average Information matrix
> This could be a problem of scale or a problem with the
> model.
> It is preferable to revise the model to remove the
> singularity.
> Specify !AISING qualifier to force the job to continue.
> Warning: 1 singularities in AI matrix.
> Source Model terms Gamma Component Comp/SE
> % C
> anim 341000 341000 0.103237 0.280705 55.44
> 63 P
> ide(anim) 341000 341000 0.120000 0.326285 0.00
> 0 S
> Variance 692840 640384 1.00000 2.71904 538.43
> 0 P
> Warning: Code B - fixed at a boundary (!GP) F - fixed by user
> ? - liable to change from P to B P - positive
> definite
> C - Constrained by user (!VCC) U - unbounded
> S - Singular Information matrix
> S means there is no information in the data for this parameter.
> Very small components with Comp/SE ratios of zero sometimes indicate
> poor
> scaling. Consider rescaling the design matrix in such
> cases.
>
> Analysis of Variance NumDF F_inc
> 6 MMC 11
> 257.14
> 15 DL 12
> 529.26
> 11 LACGET 9
> 698.30
> 4 HTD 52424 effects fitted
> 1 anim 341000 effects fitted ( 296898
> are zero)
> 18 ide(anim) 341000 effects fitted ( 296898
> are zero)
> 4355 possible outliers: see .res file
> Finished: 23 Sep 2009 17:16:56.390 Singularity in Average
> Information Matrix
>
> Right now i am re-runnig this job with !AISING
>
> How much can affect the results this Singularity in Average
> Information Matrix
> Regards,
>
> Sergio Ivan Roman Ponce
> Via Sirtori 6, 20129 Milano, Italia
> Phone 02/45490025
> Mobile +393884496900
> msn: romanponce_at_hotmail.com
> Skype: romanpsergio
> http://www.facebook.com/romanponce
>
>
> --- On Thu, 9/24/09, Arthur <asremlforum_at_VSNI.CO.UK> wrote:
>
> From: Arthur <asremlforum_at_VSNI.CO.UK>
> Subject: Re: Singularity In Average Information Matrix Using
> Ainverse
> To: ASREML-L_at_DPI.NSW.GOV.AU
> Date: Thursday, September 24, 2009, 1:53 PM
>
> Dear Michelle,
>
> It is quite straight forward.
>
> 1) You need Sire coded in the data file
>
> Then to start with, just fit Sire as a factor.
>
> Similarly, if you have Maternal Grand sire as a column in the
> data file,
>
> Add that to the model as a simple factor.
>
> That should give the main information you need to see where
> the variation lies.
>
> 2) If there is depth of pedigree (so you know parents of Sires
> and parents
> of Maternal Grand sires, you can create a single pedigree file
> giving pedigree for Sires and Maternal Grand sires.
>
> Then fit Sire and Dam using the pedigree relationship matrix.
>
>
> 2) You create a pedigree file for
>
>
>
> ______________________________________________________________
>
> Arthur Gilmour
>
> Retired Principal Research Scientist (Biometrics)
>
>
>
>
> -------------------- m2f --------------------
>
> Sent using Mail2Forum (http://www.mail2forum.com).
>
> Read this topic online here:
> http://www.vsni.co.uk/forum/viewtopic.php?p=872#872
>
> -------------------- m2f --------------------
>
>
>
>
> This message is intended for the addressee named and may contain confidential information. If you are not the intended recipient, please delete it and notify the sender. Views expressed in this message are those of the individual sender, and are not necessarily the views of their organisation.
>
>
>
>
>
> This message is intended for the addressee named and may contain confidential information. If you are not the intended recipient, please delete it and notify the sender. Views expressed in this message are those of the individual sender, and are not necessarily the views of their organisation.

-- 
Best Wishes,
Arthur Gilmour
Adjunct Professor
School of Computing and Mathematics
Charles Sturt University
Jesus went to the synagogue in every town in Galilee to preach and drive
out demons.
A leper came, knelt down and said, "If you will, you can make me clean".
Jesus was moved. He reached out and touched him and said 
"I am willing.  Be clean." He was cleansed immediately.
Mobile Number +61 427 227 468
Home phone +61 2 6364 3288  Skype: Arthur.Gilmour
http://www.CargoVale.com.au/asreml
Travel:
Adelaide AAABG 27Sept to 2 Oct
         ASReml workshop 6-8 Oct
Brisbane Probe 29Nov to 6 Dec
Bangladesh Prosihhkon 31Dec - 31 Jan