Re: Using BY statement in ASReml

Another option would be to analyse all of the data sets at once. You could
treat the data as a series of separate sites (thus with no error covariance)
and then model the variances as a series of DIAG structures.

Arthur may like to comment whether ASReml will handle estimating 1000+
variances.

I have certainly used this approach for upto 50 sites with a complex model.

Data ~ series series.fixed1 !r series.random1
500 1 1 # 500 sites, single R matrix structure for each site, 1 G sub-mtrix
structure definition
N1 0 IDEN !S2=S1 # where N1 is the number of records in series 1, with a
starting error variance of S1. Presumably they are all the same
.... #etc for the other 499 data sets

series.random1 2
series 0 DIAG R1 R2 R3 ... # starting variance R1 for random factor 1, etc
for 500 series, can put 1 variance on each line
random1 0 IDEN

Greg Dutkowski
Quantitative Geneticist
PlantPlan Genetics
c/- Locked Bag 55
School of Plant Science
University of Tasmania
GPO Hobart 7000
+61(3)62262213 Mob +61(408)238495
Fax +61(3)62262698

-----Original Message-----
From: ASReml users discussion group [mailto:ASREML-L_at_AGRIC.NSW.GOV.AU] On
Behalf Of Denis Fidalis Mujibi
Sent: Friday, 2 November 2007 8:46 AM
To: ASREML-L_at_AGRIC.NSW.GOV.AU
Subject: Re: Using BY statement in ASReml

Thanks Arthur for the help.
Given the size of the file, I have opted to go for a perl script and
the separate input files. Many regards

--
Denis Fidalis Mujibi
Ph. D. Candidate
Department of Agriculture, Food and Nutritional Science
4-10 AgroForestry Centre
University of Alberta, Edmonton
T6G2P5, AB, Canada.
fmujibi_at_ualberta.ca
Quoting arthur.gilmour_at_DPI.NSW.GOV.AU:
> Dear  Denis,
>
> Hello all,
> I was wondering if there was an equivalent statement to the SAS 'by'
> statement. This is what I mean. I have several permuted datasets that
> I have appended to each other so that the final datafile has 500
> permuted sets, each distinguised by a set number i. I would like to
> run ASReml for each i value (by i) such that I have 500 different
> outputs.
>
> THere is do  DO LOOP statment like  FOR I in 1:500
> but there are some looping facilities.
>
> There are several solutions possible
>
> Are you happy to create 500 sets of output files?
> Do you want to run a script over each set of output to capture what you
> want, rather than keeping all the files?
>
>
> In this context I would keep the data files separate in a folder  e.g.
> DATA\SET001.TXT  to  DATA\SET500.TXT
>
> and set up the job as
>
>  Title
> ...
> DATA\SET$1 ...
> ...
>
>
> and run it from a script file with command
>
> ASReml  -n job  001
> ASReml -n job 002
> etc
>
> Well, there are some many variations on the theme.
>
> The disadvantage of stacking all the data in one file  is that the whole
> file must be read reach time.
> But if you do that, then  you could use
>
> Title
> ..
> SET
> ...
> ALLDATA.TXT    !FILTER SET !SELECT $1
>
> OR ELSE
>
>
> Title
>
> ...
>  SET  !D <> $1
>
>
>
>
> Other qualifiers you can consider are  !CYCLE
>
> e.g.    ASReml -bnr job 00 01 02 03 04 05 06 07 08 09
>
> Title
> !CYCLE  1 2 3 4 5 6 7 8 9 0 !JOIN
>
> DATA\SET$1$I.txt  ..
>
> would run the jobs in 10 sets of 10   [well maybe than number the data
> sets  000 to 499]
>
>
> So when I did a big simulation,  I wrote a program when passed the data
> file to ASReml jobs
> in this way, and parsed the output.  You can do that in R or S or Fortran
> or probably perl
> and other scripting languages.
>
>
> I hope this gives some ideas.
>
>
>
> Cycling is
>
>
>
>
>
>
> May  Jesus Christ be gracious to you,
>
> Arthur Gilmour, His servant .
>
> Mixed model regression mapping for QTL detection in experimental crosses.
> Computational Statistics and Data Analysis 51:3749-3764 now available at
> http://dx.doi.org/10.1016/j.csda.2006.12.031
>
> Profile: http://www.dpi.nsw.gov.au/reader/17263
> Personal website: http://www.cargovale.com.au/
>
> mailto:Arthur.Gilmour_at_dpi.nsw.gov.au, arthur_at_cargovale.com.au
> Principal Research Scientist (Biometrics)
> NSW Department of Primary Industries
> Orange Agricultural Institute,  Forest Rd, ORANGE, 2800, AUSTRALIA
>
> fax: 02 6391 3899; 02 6391 3922                     Australia +61
> telephone work:  02 6391 3815; home: 02 6364 3288;  mobile: 0438 251 426
>
> ASREML 2 is now available from http://www.VSNi.co.uk/products/asreml
> The ASReml discussion group is at ASREML-L_at_dpi.nsw.gov.au
> To join it, mailto:arthur.gilmour_at_dpi.nsw.gov.au
> Archives are at
> https://gatekeeper.dpi.nsw.gov.au/Listserv/archives/asreml-l.html
> Cookbook: http://uncronopio.org/ASReml
>
> Proposed travel:
> UK  28 Jan
> USA 11-15 Feb
>> <><><><><><><><><><><><><><><><><><><><><><><>