Rev Language Reference
srStationarity - Stationarity stopping rule
Allow an MCMC run to terminate once the specified criterion has been met.
An MCMC sample can be considered stationary once its mean, variance and autocorrelation structure do not change over time.
Usage
Arguments
| prob : | Probability (pass by value) |
| The significance level. | |
| filename : | String (pass by value) |
| The name of the file containing the samples. | |
| frequency : | Natural (pass by value) |
| The frequency how often to check for convergence. | |
| Default : 10000 | |
| burninMethod : | String (pass by value) |
| Which type of burnin method to use. | |
| Default : ESS | |
| Options : ESS|SEM |
Details
Because the statistic is defined by comparing different runs, it can only be calculated when multiple independent runs are performed, by setting the `nruns` argument to `mcmc` or `mcmcmc` to a value greater than one.
Example
Binomial example: estimate success probability given 7 successes out of 20 trials
r ~ dnExp(10) p := Probability(ifelse(r < 1, r, 1)) n <- 20 k ~ dnBinomial(n, p) k.clamp(7) mymodel = model(k)
moves = VectorMoves() moves.append( mvSlide(r, delta=0.1, weight=1) )
paramFile = “parameters.log”
monitors = VectorMonitors() monitors.append( mnModel(filename=paramFile, printgen=100, p) )
Stop when stationarity has been attained at confidence level gamma = 0.25
stopping_rules[1] = srStationarity(prob = 0.25, file = paramFile, freq = 1000)
Create the MCMC object
mymcmc = mcmc(mymodel, monitors, moves, nruns = 2)
Begin the MCMC run
mymcmc.run(rules = stopping_rules)
Edit this page