> Hi there,
>
In the combined analysis of variance > for my unbalanced multi-year multi-location variety trials data,> I used> the following models:
(1) > Yield !WT=wt ~ mu c(trial) check check.trial !r genotype genotype.year genotype.location
> genotype.year.location !f mv
(2) Yield !WT=wt ~ mu c(trial) check check.trial date input !r genotype genotype.year genotype.location
> genotype.year.location !f mv
>
where "date" refers to the planting date of the trials and "input" is the nitrogen level of the trial.
I don't know why I had the same results from the two models. Has the factor "trial" covered all the variation caused by "date" and "input"?
>
> Your comment is greatly appreciated.
***********************************************
* Chungui Qiao *
* School of Land and Food *
* The University of Queensland *
* St Lucia, QLD 4072 *
* Phone: +61 7 3365 2859 (W) *
* Fax: +61 7 3365 1177(W) *
* Email: c.qiao@mailbox.uq.edu.au *
***********************************************
--
Asreml mailinglist archive: http://www.chiswick.anprod.csiro.au/lists/asreml