r - JAGS - different values in BUGSoutput$summary and BUGSoutput$mean -


i ran state-space model in r using r2jags package , noticed strange issue in model output. among parameters calculated model have deterministic node, is, log counts backtransformed.

nh[g,t] <- exp(lognh[g,t]) 

where g , t respectively site , time indices

in model output means reported on out$bugsoutput$mean , out$bugsoutput$summary no correspond time point on

cbind(out$bugsoutput$mean$nh[1,], out$bugsoutput$summary["nh[1,]",1])                [,1]     [,2] nh[1,1]   7.889378 7.889378 nh[1,2]   7.461558 7.461558 nh[1,3]   7.212565 7.212565 nh.....   ........ ........ nh[1,9]   5.180559 5.180559 nh[1,10]  4.987814 4.987814 nh[1,11]  5.780468 4.689696 #here starts diverging nh[1,12]  1.181960 4.311867 nh[1,13]  2.050339 3.920243 nh[1,14] 25.607913 3.693816 nh[1,15]  4.915882 3.549902 

i tried calculate manually means accessing mcmc chains (iterations=100000, burning=50000, thinnning=9, 3 chains), test output correct. example, mean of node nh[1,14] resulted

nh <- aperm(out$bugsoutput$sims.list$nh, c(2,3,1)) # simple array transposition stack iterations mean(nh[1,14,])   [1] 25.60791 

i cannot understand difference in reported means.


Comments

Popular posts from this blog

Command prompt result in label. Python 2.7 -

javascript - How do I use URL parameters to change link href on page? -

amazon web services - AWS Route53 Trying To Get Site To Resolve To www -