Evaluate Posterior Distribution of Total Population Size
This function uses the MCMC output of a "bict" object to derive an MCMC sample from the
posterior distribution of the total population size.
total_pop(object, n.burnin = 0, thin = 1, prob.level = 0.95)
object |
An object of class |
n.burnin |
An optional argument giving the number of iterations to use as burn-in. The default value is 0. |
thin |
An optional argument giving the amount of thinning to use, i.e. the computations are
based on every |
prob.level |
An optional argument giving the target probability content of the highest posterior density intervals for the total population size. The default value is 0.95. |
The use of thinning is recommended when the number of MCMC iterations and/or the number of log-linear parameters in the maximal model are/is large, which may cause problems with comuter memory storage.
This function will return an object of class "totpop" which is a list with the following components.
TOT |
A vector of length ( |
int |
The 100* |
meanTOT |
The posterior mean of the total population size. |
prob.level |
The argument |
Antony M. Overstall A.M.Overstall@soton.ac.uk.
set.seed(1) ## Set seed for reproducibility data(spina) ## Load spina data test1<-bict(formula=y~(S1+S2+S3+eth)^2,data=spina,n.sample=100,prior="UIP") ## For the spina dataset. We do 100 iterations under the unit information ## prior. The maximal model is the model with two-way interactions and we ## start from this model at the posterior model tp<-total_pop(test1,n.burnin=10) ## Use a burn-in phase of 10 iterations tp ## Print out results. Will get: #Posterior mean of total population size = 727.0667 #95 % highest posterior density interval for total population size = ( 706 757 ) ## Could do a plot ## Not run: plot(tp) ## Do a summary of MCMC sample from total population size summary(tp$TOT) ## Will get # Min. 1st Qu. Median Mean 3rd Qu. Max. # 697.0 716.2 727.0 727.1 735.8 763.0
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.