From: Rob Woolaston <Rob.Woolaston_at_BIGPOND.COM>

Date: Fri, 6 Nov 2009 06:45:05 +1000

Date: Fri, 6 Nov 2009 06:45:05 +1000

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

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.)