Re: LogL Converged, Parameter Not Converged

Hi Subash,

Why not do the following;

1. Add the !continue option from the start in your asreml parameters
file. It will be ineffective the first time but not harmful.

2. Specify a reasonably large number of iterations to start with
based on your experience with this data so far.

3. Within the script that calls the asreml program/module, write a
few codes to check the ASREMl output for specific messages using
unix/linux 'GREP' for example. There are options in python, perl etc

4. If you find this specific message then call ASREML one more time.
This time the !continue will be effective because the .rsv file will be
there and it will continue for as many iterations as originally
specified by maxit unless it converges first.

5. Depending on the number of iterations, you can repeat steps 3 and
4 once or twice.

6. If the message persist and the results are still unacceptable then
you must stop and check your data structure or follow the advice from
Bruce.

Regards

Dr. Victor E. Olori

----------------------------

Aviagen Limited

Newbridge Midlothian

EH28 8SZ

Scotland, UK.

Tel +44 (0) 131 3331056

-----Original Message-----
From: ASReml users discussion group [mailto:ASREML-L_at_DPI.NSW.GOV.AU] On
Behalf Of KRISHNAMURTHY, SUBASH [AG/1000]
Sent: 15 October 2009 17:13
To: ASREML-L_at_DPI.NSW.GOV.AU
Subject: Re: LogL Converged, Parameter Not Converged

Dear Dr. Gilmour,

I would like thank you, Dr. Cullis, Dr. Bruce, and Dr. Victor E Olori,

for your response that helped me better understand the model fitting

process in ASREML.

Here, this software is on a server within a production database. The

production database calls ASREML analysis modules (that we have written

for specific experimental designs) to analyze data from breeding trials

on a server. At this point, it is completely automated and no manual

intervention is allowed. During harvest season we have thousands of

datasets that needs to go through analysis. I agree if I were running

the analysis on my machine, I have the liberty of running it with

various different options and trying out different models. Given our

setup, I was trying to identify if we can somehow automate the process

of continuing to run, using for example, the !CONTINUE qualifier, when

the parameters do not converge, or drop that effect from the model and

fit a simpler model. I guess I can achieve this with a few programming

steps within ASREML to automate this process.

Best Regards,

-Subash

314-694-6364

-----Original Message-----

From: ASReml users discussion group [mailto:ASREML-L_at_DPI.NSW.GOV.AU] On

Behalf Of Arthur Gilmour

Sent: Wednesday, October 14, 2009 6:48 PM

To: ASREML-L_at_DPI.NSW.GOV.AU

Subject: Re: LogL Converged, Parameter Not Converged

Dear Subash,

After every iteration, ASReml rewrites the .rsv file.

If you use !CONTINUE and there is no .rsv file (the job has not been

previously run) then !CONTINUE is ignored because there is no point to

CONTINUE from.

Now if the model is changed between runs, ASReml picks up the parameters

for the parts that are recognized as the same. Further more it

recognises some progressions so the values from as a CORUV structure

may be used it initialize a XFA1 structure.

Now if using PARTs, you may have run PART 1 but not part 2.

If the .rsv file from part 1 was say SK1.rsv,

you can copy SK1.rsv to SK2.rsv

so that an .rsv file exists for part 2 and CONTINUE will work the first

time you run part 2.

ASReml 3 tries to be a little smarter and do this for you via a new .ask

file which just keeps track of which part was run last. Then if you

run part 1, then part 2 with continue, ASReml will check for SK2.rsv

and since it is not there, identify SK1.rsv as being the .rsv file that

should be used.

The whole aim is to avoid the tedious and error prone process of

manually typing in revised initial values after running an initial job.

Best wishes.

On Wed, 2009-10-14 at 09:11 -0500, KRISHNAMURTHY, SUBASH [AG/1000]

wrote:

> I am still trying to understand how this !CONTINUE works and I

> apologize, if in the process, I am asking the same questions over and

> over.

>

> It seems to me that for the qualifier !CONTINUE to work, the .rsv file

> must be sitting there. Does this mean that I have to run the program

> twice each time I see this message of Parameter Not Converged? If

> that is the case, is there a way to tell ASREML to do this

> automatically.

>

> Also, is it safe to use the outputs like the Least Squared Means

> (BLUPS), standared erros of the LSM (BLUPS) etc., with this warning

> message?

>

>

-- 
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
 
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 e-mail message may contain privileged and/or confidential
information, and is intended to be received only by persons entitled to
receive such information. If you have received this e-mail in error,
please notify the sender immediately. Please delete it and all
attachments from any servers, hard drives or any other media. Other use
of this e-mail by you is strictly prohibited.
 
 
All e-mails and attachments sent and received are subject to monitoring,
reading and archival by Monsanto, including its subsidiaries. The
recipient of this e-mail is solely responsible for checking for the
presence of "Viruses" or other "Malware". Monsanto, along with its
subsidiaries, accepts no liability for any damage caused by any such
code transmitted by or accompanying this e-mail or any attachment.
------------------------------------------------------------------------
---------------------------------
 
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 e-mail contains confidential information and is intended solely for use by the individual named or entity to whom it is addressed.  Please notify the sender and postmaster_at_aviagen.com immediately by e-mail if you have received this e-mail by mistake and delete this e-mail from your system.  If you are not the named addressee, you are notified that disclosing, copying, distributing or taking any action in reliance on the contents of this information is strictly prohibited.  Aviagen accepts no liability for the content of this e-mail and any consequences of actions taken in reliance on the contents, unless that information is subsequently confirmed in writing.  Any views, opinions or actions presented in this e-mail are solely the author's and have not been approved by Aviagen.  Any defamatory statements or infringing communication is contrary to Aviagen policy and outside the scope of the employment of the individual concerned.  No employee or agent is authorized to conclude any binding agreement on behalf of Aviagen with another party by e-mail. Aviagen has taken reasonable precautions to ensure no viruses are present in this e-mail, but cannot accept responsibility for any loss or damage arising from the use of this e-mail or attachments.
*****************************************************************************