Re: Different Models producing same results

Subash

This suggests to me that your entries are coded in such a way that once you
have specified entry, you have automatically specified group. No matter
what you put on the model line, ASReml can work out that entries are in fact
nested within group and so the two factors cannot be independent. If the
same entry code appeared in more than one group, the degrees of freedom
would be different in the two models and you'd expect different answers. If
you had "ENTRY" spelt correctly on the tabulate line, this would become
apparent in the .tab file.

rw

  _____

From: ASReml users discussion group [mailto:ASREML-L_at_DPI.NSW.GOV.AU] On
Behalf Of KRISHNAMURTHY, SUBASH [AG/1000]
Sent: Friday, 6 November 2009 6:04 AM
To: ASREML-L_at_DPI.NSW.GOV.AU
Subject: Different Models producing same results

All,

I am fitting the following two models in ASREML. They are identical except
for the part highlighted in YELLOW. The two models produce identical
results. The first model is saying ENTRY is nested within GROUP1 and in the
second model, the two factors GROUP1 and ENTRY are modeled as independent
factors. Why would the two models produce identical results?

Model1

!PATH 4 #MULTI LOC MULTI REP BLUE G1-Fixed ENTRY-FIXED

tabulate RESPONSE ~ GROUP1 ENRTY

RESPONSE ~ mu GROUP1/ENTRY !r LOC/REP LOC.GROUP1 LOC.REP.GROUP1

predict GROUP1 !PRESENT GROUP1 ENTRY !SED !TDIFF

predict GROUP1.ENTRY !SED !TDIFF

Model2

!PATH 4 #MULTI LOC MULTI REP BLUE G1-Fixed ENTRY-FIXED

tabulate RESPONSE ~ GROUP1 ENRTY

RESPONSE ~ mu GROUP1 ENTRY !r LOC/REP LOC.GROUP1 LOC.REP.GROUP1

predict GROUP1 !PRESENT GROUP1 ENTRY !SED !TDIFF

predict GROUP1.ENTRY !SED !TDIFF

ASR file from Model1

ASR File from Model 2

ASReml 2.00b [01 Jul 2006] ASREML GUBD MODULE

     Build: w [ 1 Oct 2008] 32 bit

 05 Nov 2009 13:52:27.004 1264.00 Mbyte Windows GUBD_Integrated_Module

 Licensed to: Monsanto

 ***********************************************************

 * SYNTAX change: A/B now means A A.B *

 * *

 * Contact support_at_asreml.co.uk for licensing and support *

 ***************************************************** ARG *

           Approximate stratum variance decomposition

 Stratum Degrees-Freedom Variance Component Coefficients

 LOC 3.00 834.703 18.4 9.1 1.8 1.0

 LOC.REP 4.02 441.324 0.0 9.7 1.9 1.0

 LOC.REP.GROUP1 27.93 496.102 0.0 0.0 3.6 1.0

 Residual Variance 131.05 29.3327 0.0 0.0 0.0 1.0

 Source Model terms Gamma Component Comp/SE % C

 LOC 4 4 0.771952 22.6434 0.56 0 P

 LOC.REP 8 8 0.606630 17.7941 0.54 0 P

 LOC.GROUP1 20 20 0.535386E-07 0.157043E-05 0.00 0 B

 LOC.REP.GROUP1 40 40 4.38474 128.616 3.51 0 P

 Variance 192 166 1.00000 29.3327 8.09 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 DenDF_con F_inc F_con M P_con

  10 mu 1 3.0 2786.24 2786.24 .
<.001

   5 GROUP1 4 28.2 1.48 1.48 A 0.235

  11 GROUP1.ENTRY 21 131.7 1.06 1.06 B 0.405

 Notice: The DenDF values are calculated ignoring fixed/boundary/singular

             variance parameters using algebraic derivatives.

   1 LOC 4 effects fitted

  12 LOC.REP 8 effects fitted

  13 LOC.GROUP1 20 effects fitted

  15 LOC.REP.GROUP1 40 effects fitted

 Finished: 05 Nov 2009 13:52:27.129 LogL Converged

ASReml 2.00b [01 Jul 2006] ASREML GUBD MODULE

     Build: w [ 1 Oct 2008] 32 bit

 05 Nov 2009 13:53:00.399 1264.00 Mbyte Windows GUBD_Integrated_Module

 Licensed to: Monsanto

 ***********************************************************

 * SYNTAX change: A/B now means A A.B *

 * *

 * Contact support_at_asreml.co.uk for licensing and support *

 ***************************************************** ARG *

          Approximate stratum variance decomposition

 Stratum Degrees-Freedom Variance Component Coefficients

 LOC 3.00 834.703 18.4 9.1 1.8 1.0

 LOC.REP 4.02 441.324 0.0 9.7 1.9 1.0

 LOC.REP.GROUP1 27.93 496.102 0.0 0.0 3.6 1.0

 Residual Variance 131.05 29.3327 0.0 0.0 0.0 1.0

 Source Model terms Gamma Component Comp/SE % C

 LOC 4 4 0.771952 22.6434 0.56 0 P

 LOC.REP 8 8 0.606630 17.7941 0.54 0 P

 LOC.GROUP1 20 20 0.535386E-07 0.157043E-05 0.00 0 B

 LOC.REP.GROUP1 40 40 4.38474 128.616 3.51 0 P

 Variance 192 166 1.00000 29.3327 8.09 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 DenDF_con F_inc F_con M P_con

  10 mu 1 3.0 2786.24 2786.24 .
<.001

   5 GROUP1 4 28.2 1.48 1.48 a 0.235

   8 ENTRY 21 131.7 1.06 1.06 A 0.405

 Notice: The DenDF values are calculated ignoring fixed/boundary/singular

             variance parameters using algebraic derivatives.

   1 LOC 4 effects fitted

  11 LOC.REP 8 effects fitted

  12 LOC.GROUP1 20 effects fitted

  14 LOC.REP.GROUP1 40 effects fitted

 Finished: 05 Nov 2009 13:53:00.477 LogL Converged

Best Regards,

Subash Krishnamurthy

BioStatistician

Monsanto

ADR: 800 North Lindbergh Blvd.

Creve Ceour MO-63376

TEL: 314-694-6364

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.

 
Received on Sat Nov 06 2009 - 06:45:05 EST