variance partitioning: vegan
It uses the number of columns in the *model* matrix. Factors get expanded into contrasts in the model matrix: a p-level factor will be expanded into p-1 columns in the model matrix. Cheers, Jari Oksanen
From: r-sig-ecology-bounces at r-project.org <r-sig-ecology-bounces at r-project.org> on behalf of Martin Weiser <weiser2 at natur.cuni.cz>
Sent: 23 October 2014 15:28
To: R-sig-ecology at r-project.org
Subject: [R-sig-eco] variance partitioning: vegan
Sent: 23 October 2014 15:28
To: R-sig-ecology at r-project.org
Subject: [R-sig-eco] variance partitioning: vegan
Dear friends, I have just looked at vegan:::varpart2, and it seems to me that in estimation of degrees of freedom for adjusted R2 calculation, it just uses number of columns in the constraining matrix/dataframe. Am I right? Is this appropriate if the constraining dataframe involves factors with several levels? Best, Martin W. -- ------------------------------ Pokud je tento e-mail sou??st? obchodn?ho jedn?n?, P??rodov?deck? fakulta Univerzity Karlovy v Praze: a) si vyhrazuje pr?vo jedn?n? kdykoliv ukon?it a to i bez uveden? d?vodu, b) stanovuje, ?e smlouva mus? m?t p?semnou formu, c) vylu?uje p?ijet? nab?dky s dodatkem ?i odchylkou, d) stanovuje, ?e smlouva je uzav?ena teprve v?slovn?m dosa?en?m shody na v?ech n?le?itostech smlouvy. _______________________________________________ R-sig-ecology mailing list R-sig-ecology at r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-ecology