Re: Multivariate analysis using a giv file

Dear Arthur,

Many thanks for your prompt help. After modifying .as file, the ASreml worked. However, I have noticed that the results are for part 1 only, not parts 2 and 3.
Please see the following printout.

 ASReml 2.00b [01 Jul 2006] gcma pond and raceway animal studies
     Build: w [ 1 Oct 2008] 32 bit
 16 Jan 2009 13:40:59.967 32.00 Mbyte Windows ffpondrace1
 Licensed to: CSIRO
 ***********************************************************
 * SYNTAX change: A/B now means A A.B *
 * *
 * Contact support_at_asreml.co.uk for licensing and support *
 ***************************************************** ARG *
 Folder: C:\Data\Data\Backed Up Folder\AFT5-FFF\New Microsatellites-Industry\GCMA 2007 test\comparison\female
  animal !I 576
  site 4 !A
  sx 1 !A
  Reading ped.giv skipping 0 header lines
 Inverse G structure of 576 rows having 166176 non zero cells read from ped.giv
 QUALIFIERS: !SKIP 1
 QUALIFIERS: !FILTER environ !SELECT 1
 QUALIFIER: !DOPART 1 is active
 Reading ffphen1.dat FREE FORMAT skipping 1 lines

 Univariate analysis of weight
 Using 480 records of 576 read

  Model term Size #miss #zero MinNon0 Mean MaxNon0
   1 animal 576 0 0 1 240.5000 480
  Warning: Fewer levels found in environ than specified
   2 environ 2 0 0 1 1.0000 1
   3 site 4 0 0 1 1.5104 2
  Warning: Fewer levels found in sx than specified
   4 sx 2 0 0 1 1.0000 1
   5 weight Variate 0 0 5.700 40.91 65.10
   6 mu 1
   7 giv(animal,1) 576 0 0 1 240.5000 480
 Forming 581 equations: 5 dense.
 Initial updates will be shrunk by factor 0.316
 Notice: Algebraic ANOVA Denominator DF calculation is not available
         Numerical derivatives will be used.

 ** Notice **: GIV matrix not positive definite: Negative pivots 261 576

 Notice: 3 singularities detected in design matrix.
   1 LogL=-1564.27 S2= 152.10 478 df 0.2500 1.000
   2 LogL=-1563.59 S2= 150.17 478 df 0.2848 1.000
   3 LogL=-1563.07 S2= 147.98 478 df 0.3337 1.000
   4 LogL=-1562.97 S2= 146.56 478 df 0.3740 1.000
   5 LogL=-1562.97 S2= 146.82 478 df 0.3660 1.000
   6 LogL=-1562.97 S2= 146.75 478 df 0.3680 1.000
 Final parameter values 0.36750 1.0000

          Approximate stratum variance decomposition
 Stratum Degrees-Freedom Variance Component Coefficients
 giv(animal,1) 0.00 0.00000 0.0 1.0
 Residual Variance 441.23 146.766 0.0 1.0

 Source Model terms Gamma Component Comp/SE % C
 giv(animal,1) 576 576 0.367622 53.9544 4.56 0 P
 Variance 480 478 1.00000 146.766 14.85 0 P

 Analysis of Variance NumDF DenDF F_inc Prob
   6 mu 1 63.5 116.43 <.001
   3 site 1 461.1 0.41 0.523
 Notice: The DenDF values are calculated ignoring fixed/boundary/singular
             variance parameters using numerical derivatives.

                     Estimate Standard Error T-value T-prev
   3 site
     pond3 0.749537 1.16725 0.64
   6 mu
                    1 40.7343 3.84461 10.60
   7 giv(animal,1) 576 effects fitted
 Finished: 16 Jan 2009 13:41:19.888 LogL Converged
==============================================

Does this mean the animal orders in GIV file is not right?

Thanks in advance

Yutao Li

-------------------- m2f --------------------

Sent using Mail2Forum (http://www.mail2forum.com).

Read this topic online here:
http://www.vsni.co.uk/forum/viewtopic.php?p=352#352

-------------------- m2f --------------------
Received on Sat Jan 16 2009 - 05:35:59 EST