[R] Question about R graphs
legendy
junli at uab.edu
Wed Oct 22 20:34:01 CEST 2008
Hello all,
I have a question about CODA. Using my R code as bellow, I drew two graphs
(the data comes from the example of the speed of light in the Gelman’s book
“Bayesian Data Analysis (Second edition)” (page 77)). My R code is:
library(R2WinBUGS)
y=c(28, 26, 33, 24, 34, -44, 27, 16, 40, -2, 29, 22, 24, 21, 25, 30, 23, 29,
31,
19, 24, 20, 36, 32, 36, 28, 25, 21, 28, 29, 37, 25, 28, 26, 30, 32, 36,
26,
30, 22, 36, 23, 27, 27, 28, 27, 31, 27, 26, 33, 26, 32, 32, 24, 39, 28,
24,
25, 32, 25, 29, 27, 28, 29, 16, 23)
n=length(y)
data=list("n","y")
inits=function() {list(mu=runif(1,-5,5),tau=runif(1,0.01,10))}
parameters=c("mu","sigma")
sim=bugs(data,inits,parameters,"U:/my BUGS
code.txt",n.chains=3,n.iter=1000,debug=T)
library(coda)
y=sim$sims.array
x1=mcmc(y[,1,]) #mu,sigma, and deviance from the chain 1
x2=mcmc(y[,2,]) #mu,sigma, and deviance from the chain 2
x3=mcmc(y[,3,]) #mu,sigma, and deviance from the chain 3
x=mcmc.list(x1,x2,x3)
plot(x)
gelman.plot(x)
# my BUGS code is (copy and save as a txt file):
model
{
for(i in 1:n)
{
y[i]~dnorm(mu,tau)
}
mu~dnorm(0.0,0.001)
tau~dgamma(0.001,0.001)
sigma<-1/sqrt(tau)
}
As you see, I set 3 chains and 1000 iterations, so there should be 1500
simulation samples after removed the first 500 samples for each chain (burn
in). But from the density graph, there indicated only N=500. Why N=500? But
not 1500.
>From another graph produced by “gelman.plot”, we can see that the x axis is
labeled as “last iteration in chain”. What is the last iteration in chain? I
set 3 chains, which chain was the result calculated based on?
Thank you very much.
Legendy
--
View this message in context: http://www.nabble.com/Question-about-R-graphs-tp20117061p20117061.html
Sent from the R help mailing list archive at Nabble.com.
More information about the R-help
mailing list