Dear Statistical/Biometrical colleagues,
This note is to inform you of the existence of ASREML,
to give you the opportunity to trial it,
and to advise you of a workshop on its use.
WHAT IS ASREML?
ASREML is a program designed to fit the general linear mixed model
to large data sets using complex variance models.
Its operational efficiency arises from using
the Average Information REML procedure (giving quadratic convergence)
and Sparse matrix operations to handle large models.
Problem size depends on the sparsity of the mixed model equations
and the size of your computer but models with 100,000 effects
have been fitted successfully.
ASREML was developed specifically for modeling complex variance structures
and has application in
Spatial Analysis of field experiments,
Meta analysis of trials with common treatments (both simultaneous
analysis of several trials with common treatments and 2 stage
analysis of the means from many trials)
Repeated measures analysis ( Manova type models or Spline type models)
Genetic analysis (involving a numerator relationship matrix)
of univariate and multivariate data.
ASREML has been operational for 2 years and is in a stable form although
continual development is planned to extend the variance modeling options.
WHERE CAN I GET IT?
An evaluation copy of ASREML for Sun Solaris and Windows 95/NT PCs
is available from the address below. A postscript manual of 130 pages
is also on the ftp site. For the time being, no charge is made but
this will change soon.
HOW EASY IS IT TO USE?
ASREML has a powerful but terse user interface as most effort has been
directed towards efficiency of the 'core'. It operates purely in a BATCH
mode. The core of ASREML forms the basis of the new REML procedure in
Genstat and David Butler is developing an interface for Splus. However,
some users will need to use ASREML itself because these other interfaces
will have additional overheads.
ASREML USER'S WORKSHOP
A workshop for ASREML users will be conducted by Arthur GILMOUR at Orange
on 10-12 March from 9.00 am to 4.00 pm each day.
Tues 10th March Formal presentation of statistical analysis in ASREML.
Wed 11th March Tutorial - bring your own data to analyse
Thurs 12th March Tutorial - bring your own data to analyse
Participants can bring their own PC or use ASREML on our SUN.
The cost is $300 ($100 for students)
plus your travel and accommodation expenses.
To register, please email me. You may bring the registration fee
or NSW Agriculture will invoice you after the event.
Attendance is limited to 20 participants.
REGISTRATION FORM (Please email the following information to
Kinds of data you would like to analyses in ASREML:
Registration type (Full/Student:Institution)
Arthur Gilmour PhD email: Arthur.Gilmour@agric.nsw.gov.au
Senior Research Scientist (Biometrics) fax: <61> 2 6391 3899
NSW Agriculture telephone work: <61> 2 6391 3815
Orange Agricultural Institute home: <61> 2 6362 0046
Forest Rd, ORANGE, 2800, AUSTRALIA
ASREML is currently free by anonymous ftp from pub/aar on ftp.res.bbsrc.ac.uk
Point your web browser at ftp://ftp.res.bbsrc.ac.uk/pub/aar/
in the IACR-Rothamsted information system http://www.res.bbsrc.ac.uk/
To join the asreml discussion list, send the message
The address for messages to the list is firstname.lastname@example.org.CSIRO.au
>>>>>>>>>>>> ASREML WORKSHOP scheduled 10-12 March: $300 <<<<<<<<<<<<<
But He answered and said, "It is written, `Man shall not live by bread alone,
but by every word that proceeds from the mouth of God' " Matthew 4:4