Variation Partitioning For GLM Or GAM
Perform variance partitioning for binomial GLM or GAM based on the deviance of two groups or predicting variables.
ecospat.varpart (model.1, model.2, model.12)
model.1 |
GLM / GAM calibrated on the first group of variables. |
model.2 |
GLM / GAM calibrated on the second group of variables. |
model.12 |
GLM / GAM calibrated on all variables from the two groups. |
The deviance is calculated with the adjusted geometric mean squared improvement rescaled for a maximum of 1.
Return the four fractions of deviance as in Randin et al. 2009: partial deviance of model 1 and 2, joined deviance and unexplained deviance.
Christophe Randin christophe.randin@unibas.ch, Helene Jaccard and Nigel Gilles Yoccoz
Randin, C.F., H. Jaccard, P. Vittoz, N.G. Yoccoz and A. Guisan. 2009. Land use improves spatial predictions of mountain plant abundance but not presence-absence. Journal of Vegetation Science, 20, 996-1008.
library(rms)
data('ecospat.testData')
# data for Soldanella alpina and Achillea millefolium
data.Solalp<- ecospat.testData[c("Soldanella_alpina","ddeg","mind","srad","slp","topo")]
# glm models for Soldanella alpina
glm.Solalp1 <- glm("Soldanella_alpina ~ pol(ddeg,2) + pol(mind,2) + pol(srad,2)",
data = data.Solalp, family = binomial)
glm.Solalp2 <- glm("Soldanella_alpina ~ pol(slp,2) + pol(topo,2)",
data = data.Solalp, family = binomial)
ecospat.varpart (model.1= glm.Solalp1, model.2= glm.Solalp2, model.12= glm.Solalp2)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.