I am analyzing some data that came from demographic health surveys. The
data contain information for individuals within households, that are
located within clusters, that are located within survey years, that are
located within countries. We are trying to find the best model from a
subset of predictors, and all models must contain the random variable of
household within cluster within year within country.
We are running models on a server with 64GB memory and 6 CPU cores
Data are available at:
First, we tested a linear mixed model using the lmer package:
########## Install Packages ##########
library(lme4)
library(glmulti)
########## Clear all memory/objects ##########
rm(list=ls())
########## Read in Data ##########
mydata = read.csv("kr.and.GIS.cleaned.Residents.only.csv")
###################################################################
########## Try lmer model NO Interactions ##########
###################################################################
ptm <- proc.time()
lmer.model = lmer(stunt.dhs ~
dis_ed_des+
tc_pa+
avg_clu_tc+
dist_road+
pden_lscan+
URBAN_RURA+
time.water+
wealth.index +
(1|country.code.short/year/cluster/household),
mydata,
REML = F)
print(lmer.model)
proc.time() - ptm
This works fine and gives the following output
--
Alicia Ellis
Postdoc
Gund Institute for Ecological Economics
University of Vermont
617 Main Street
Burlington, VT 05405
(802) 656-1046
http://www.wcs-heal.org
http://www.uvm.edu/~aellis5
<http://entomology.ucdavis.edu/faculty/scott/aellis/>
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org 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.
