[R] The 'REP' term in AMMI{agricolae}

[joris.dewolf at cropdesign.com]

The AMMI senso strictu part starts from the corrected data for genotype and
environment. In most cases where AMMI is applied (maybe also in the
agricolae package, I haven't checked), starts from the interaction effects
obtained through a general linear model anova.
It should be possible to replace the input by blups from your mixed model.
R news 7/1 (p14-19) gave some basic code for AMMI. There you should be able
to get ideas how to modify the code.


Joris








                                                                           
             "CG Pettersson"                                               
             <cg.pettersson at vp                                             
             e.slu.se>                                                  To 
             Sent by:                  r-help at stat.math.ethz.ch            
             r-help-bounces at st                                          cc 
             at.math.ethz.ch                                               
                                                                   Subject 
                                       [R] The 'REP' term in               
             23/07/2007 19:59          AMMI{agricolae}                     
                                                                           
                                                                           
                                                                           
                                                                           
                                                                           
                                                                           




Dear all,

W2k, R 2.5.1

I am trying out the AMMI function in the agricolae package, to analyse the
dependence of environment for a certain cultivar. The function responds to
four basic variables:

ENV Environment
GEN Genotype
REP Replication
Y Response

My question concerns the 'REP' term. When I normally do an analysis of
variance, the replication would refer to repeated observations within the
each field trial. In the case I am analysing right now, I have five years
of observations for each 'ENV', in such a case it feels natural to use
year as the most important replication of the environment- especially as I
am interested in long term trends. This approach also seems to work
allright.

But the field trials are also structured in three main blocks, subdivided
into five 'lattice' blocks, a structure that is powerful in the analysis
of variance. (I use a random call in lme{nlme}).

Is it possible to use the block structure also in the AMMI analysis? If
that is possible, how should I code? I have tried to find out in the
documentation, but if it is stated there I do not understand it.

Thank you
/CG

--
CG Pettersson, PhD
Swedish University of Agricultural Sciences (SLU)
Dept. of Crop Production Ecology. Box 7043.
SE-750 07 Uppsala, Sweden
cg.pettersson at vpe.slu.se

______________________________________________
R-help at stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.