Re: LogL Converged, Parameter Not Converged

From: KRISHNAMURTHY, SUBASH [AG/1000] <subash.krishnamurthy_at_MONSANTO.COM>
Date: Thu, 15 Oct 2009 11:12:48 -0500

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,


-----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
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

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]
> 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
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.
Received on Fri Oct 15 2009 - 11:12:48 EST

This webpage is part of the ASReml-l discussion list archives 2004-2010. More information on ASReml can be found at the VSN website. This discussion list is now deprecated - please use the VSN forum for discussion on ASReml. (These online archives were generated using the hypermail package.)