Fixed effects problem

[Leigh.Callinan@nre.vic.gov.au]

Dear ASREML users,

Can anyone see from the AS & ASR files below why I can't get any main effects
from ASREML analysis of variance. There's about 30 units in each of the 6
factor combinations. I've never had a factorial model do this before. I've
tried quite a few possible solutions with no luck.



Analysis of lysimeter data for Christie Ho ho193.as
 core 24
 wtmm  3 !A
 defmm  2 !A
 days
 period 8 !A
 infil !^0
 dm252
 dm303
 dn274
ho193.prn !skip 1
infil ~  mu mv defmm wtmm defmm.wtmm   # !r core  + period
# 1 2
#1 core IDEN
#56 period ARMA .5
#0
#predict wtmm
#predict defmm




 *********************************************** ARG *
 QUALIFIERS: !SKIP 1
  Reading ho193.prn  FREE FORMAT skipping  1  lines
 Univariate analysis of infil
 Using      184 records [of     184 read from     184 lines of ho193.prn
           ]
  Model term      Size Type    COL   Minimum    Mean      Maximum   #zero #miss
   1 core           24 Factor    1      1    12.4022         24         0     0
   2 wtmm            3 Factor    2      1     1.9891          3         0     0
   3 defmm           2 Factor    3      1     1.5217          2         0     0
   4 days            1 Covariat  4 0.7000E-02  6.887      29.00         0     0
   5 period          8 Factor    5      1     4.3967          8         0     0
   6 infil           1 Variate   6 0.4055      3.646      4.934         0     1
   7 dm252           1 Covariat  7 0.4495     0.9062      1.308         0     0
   8 dm303           1 Covariat  8 0.4398     0.8368      1.277         0     0
   9 dn274           1 Covariat  9 0.4185     0.6947      1.140         0     0
  10 mu              1 Constant Term
  11 mv_estimates    1 Missing value
  12 defmm.wtmm      6 Interaction  3 defmm   :    2    2 wtmm           :    3
  Forming  13  equations:  13  dense
 Initial updates will be shrunk by factor    0.316
 NOTICE:     6 (more) singularities,
 LogL=-34.8413     S2= 0.48572        177 df    1.000
 Final parameter values                        1.0000

 Source                Model  terms     Gamma     Component    Comp/SE   % C
 Variance                184    177   1.00000      0.485718       9.41   0 P

 Analysis of Variance       DF     F-incr      F-adj  StndErrDiff
  12 defmm.wtmm              6     835.96     840.41
  11 mv_estimates            1      26.72      26.72

                     Solution       Standard Error    T-value     T-prev
  11 mv_estimates
                    2   -3.66028       0.708087         -5.17
  12 defmm.wtmm
                    8    3.33282       0.125173         26.63
                    9    3.57048       0.131708         27.11      1.31
                   10    3.23636       0.129418         25.01     -1.81
                   11    4.11401       0.123202         33.39      4.91
                   12    3.90213       0.123202         31.67     -1.22
                   13    3.66028       0.125173         29.24     -1.38
 3  possible outliers: see .res file
 Finished: 19 Mar 2001 12:41:58.523   LogL Converged


--
Asreml mailinglist archive: http://www.chiswick.anprod.csiro.au/lists/asreml