Re[2]: using predict statement
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re[2]: using predict statement



Hi Arthur,

Yes, the mothers are nested in subraces. There are more subraces (22
in total) for other traits like stem diameter, but I had only
information for 8 with PPY. When trying to use:

predict subrace !present mother subrace

I got the following error:
jwe0019i-u The program was terminated abnormally with Exception Code
EXCEPTION_ACCESS_VIOLATION

Error occurs at or near line 623 of _g5vprd_
Called from or near line 99 of _aiwork_
Called from or near line 2896 of _aidata_
Called from or near line 60 of _MAIN__
Error summary (Fortran)
error number error level error condition
jwe0019i     u           i
total error count 1

Best regards,


Luis

-- 
Dr Luis A. Apiolaza
Quantitative Geneticist
CRC for Sustainable Production Forestry
School of Plant Science
University of Tasmania
GPO Box 252-55
Hobart TAS 7001
Australia

phone:   +61-3-6226 2213
fax:     +61-3-9229 2698
email:   mailto:luis.apiolaza@utas.edu.au

'We can only see a short distance ahead, but we can
 see plenty there that needs to be done'--Alan Turing
http://www.scieng.utas.edu.au/plantsci/
http://www.geocities.com/uncronopio/
;-)

On 12/20/2001 you wrote:


>> 
>> Hi,
>> 
>> I'm starting to use the predict statement, but I'm not sure about how
>> does it work. Using a model like:
>> 
>> !part 13
>> PPY ~ mu subrace rep !r mother
>> 0
>> predict subrace
>> 
>> I obtain results for subraces like:
>>  rep                  is averaged over fixed levels
>>  Predicted values of PPY
>>  mother               is ignored in the prediction

Arthur> Note the above comment:  You would get different (smaller) standard errors if
Arthur> the prediction averaged over mother by saying
Arthur>   predict subrace !ave mother
  
Arthur>   .  If mother was nested within subrace, you would need to say
Arthur>   predict subrace !present mother subrace
  
>>  WARNING:  Missing cells in table
Arthur>      Can you see what is missing? were there more subraces?

>> 
>>  subrace             Predicted_Value Standard_Error EPcode                                   
>>  "Flinders Island"           51.1017         0.4593 P  
>>  "King Island"               51.7799         0.4051 P  
>>  "Strzelecki Ranges"         51.2644         0.4696 P  
>>  "Southern Tasmania"         52.0033         0.4782 P  
>>  "Western Otways"            52.2484         0.4108 P  
>>  "North-eastern Tasma        50.5914         0.6675 P
>>  "South-eastern Tasma        51.9727         0.3643 P  
>>  "Strzelecki Foothill        50.7930         0.4653 P  
>>  Standard Error of Difference: Min     0.5433   Mean     0.6613   Max     
Arthur> 0.8188
>> 
>> From the as file mu is 50.4851 with sdterr 0.490574. Shouldn't some of
>> the subrace solutions be over and some under the overall mean? That is
>> the case if I calculate mu+subrace effects (either from the .asr or
>> the .sln files). I would appreciate an explanation about how does
>> predict work.

Arthur>   Leave  mu  out of the model and see if you get the same prediction.
Arthur>   Averaging over rep is probably causing the confusion.  The fitted mu
Arthur>   relates to a rep_1 effect of zero  while the predicted values
Arthur>   relate to an average rep effect of zero.  i.e.  the difference between what 
Arthur> you calculated
Arthur>   from the .sln file and what asreml gives in the pvs file should be
Arthur>   the average rep effect.
  
Arthur>   Arthur
>> 
>> Best regards,
>> 
>> 
>> Luis
>> 
>> -- 
>> Dr Luis A. Apiolaza
>> Quantitative Geneticist
>> CRC for Sustainable Production Forestry
>> School of Plant Science
>> University of Tasmania
>> GPO Box 252-55
>> Hobart TAS 7001
>> Australia
>> 
>> phone:   +61-3-6226 2213
>> fax:     +61-3-9229 2698
>> email:   mailto:luis.apiolaza@utas.edu.au
>> 
>> 'The search for truth is more precious than
>>  its possession' --Albert Einstein
>> http://www.scieng.utas.edu.au/plantsci/
>> http://www.geocities.com/uncronopio/
>> ;-)
>> 
>> --
>> Asreml mailinglist archive: http://www.chiswick.anprod.csiro.au/lists/asreml


Arthur> <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
Arthur> Arthur Gilmour PhD                  mailto:Arthur.Gilmour@agric.nsw.gov.au
Arthur> Principal Research Scientist (Biometrics)            fax: <61> 2 6391 3899
Arthur> NSW Agriculture                                           <61> 2 6391 3922
Arthur> Orange Agricultural Institute             telephone work: <61> 2 6391 3815
Arthur> Forest Rd, ORANGE, 2800, AUSTRALIA                  home: <61> 2 6364 3288
Arthur>                                                   mobile: <61> 4 1964 7675
Arthur> ASREML is available by anonymous ftp from pub/aar on ftp.res.bbsrc.ac.uk
Arthur>     or point your web browser at ftp://ftp.res.bbsrc.ac.uk/pub/aar/ 

Arthur> To join the asreml discussion list, send the message  
Arthur>      subscribe
Arthur> mailto:asreml-request@chiswick.anprod.CSIRO.au

Arthur> To send messages to the list, mailto:asreml@chiswick.anprod.CSIRO.au

Arthur> Asreml list archive: http://www.chiswick.anprod.csiro.au/lists/asreml

Arthur>                         <> <> <> <> <> <> <>
Arthur> "Blessed are the peacemakers,
Arthur>         for they shall be called sons of God"   Jesus; Matthew 5:9

Arthur> <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

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