• File: test4.cmd
/* SEEDS example with uniform prior on the standard deviation of the
   random effects. The autocorrelation is smaller than for other 
   parameterizations, allowing a shorter run.
*/
model in "seedsuni.bug" 
data in "seeds-data.R"
load glm
compile, nchains(2)
inits in "seeds-init.R"
initialize
update 5000 
monitor alpha0 , thin(10)
monitor alpha1 , thin(10)
monitor alpha2 , thin(10)
monitor alpha12 , thin(10)
monitor sigma , thin(10)
update 10000 
coda *