Rev Language Reference
mcmc - MCMC analysis object
The MCMC analysis object keeps a model and the associated moves and monitors. The object is used to run Markov chain Monte Carlo (MCMC) simulation on the model, using the provided moves, to obtain a sample of the posterior probability distribution. During the analysis, the monitors are responsible for sampling model parameters of interest.
Usage
mcmc(Model model, Monitor[] monitors, Move[] moves, String moveschedule, Natural nruns, String combine, Natural ntries, RealPos priorHeat, RealPos likelihoodHeat, RealPos posteriorHeat)
Arguments
| model : | Model (pass by value) |
| The model graph. | |
| monitors : | Monitor[] (pass by value) |
| The monitors used for this analysis. | |
| moves : | Move[] (pass by value) |
| The moves used for this analysis. | |
| moveschedule : | String (pass by value) |
| The strategy how the moves are used. | |
| Default : random | |
| Options : sequential|random|single | |
| nruns : | Natural (pass by value) |
| The number of replicate analyses. | |
| Default : 1 | |
| combine : | String (pass by value) |
| How should we combine the traces once the simulation is finished. | |
| Default : none | |
| Options : sequential|mixed|none | |
| ntries : | Natural (pass by value) |
| The number of initialization attempts. | |
| Default : 1000 | |
| priorHeat : | RealPos (pass by value) |
| The power that the prior will be raised to. | |
| Default : 1 | |
| likelihoodHeat : | RealPos (pass by value) |
| The power that the likelihood will be raised to. | |
| Default : 1 | |
| posteriorHeat : | RealPos (pass by value) |
| The power that the posterior will be raised to. | |
| Default : 1 |
Details
The MCMC analysis object produced by a call to this function keeps copies of the model and the associated moves and monitors. The MCMC analysis object is used to run Markov chain Monte Carlo (MCMC) simulation on the model, using the provided moves, to obtain a sample of the posterior probability distribution. During the analysis, the monitors are responsible for sampling model parameters of interest.
Example
# Create a simple model (unclamped)
a ~ exponential(1)
mymodel = model(a)
# Create a move vector and a monitor vector
moves[1] = mvScale(a, lambda=1.0, weight=1.0)
monitors[1] = mnFile(a,"output/out.log")
# Create an mcmc object
mymcmcObject = mcmc( mymodel, monitors, moves)
# Run a short analysis
mymcmcObject.burnin( generations = 400, tuningInterval = 100)
mymcmcObject.run( generations = 400)
# print the summary of the operators (now tuned)
mymcmcObject.operatorSummary()
Methods
- burnin(Natural generations, Natural tuningInterval, Bool underPrior)
- initializeFromCheckpoint(String checkpointFile)
- initializeFromTrace(ModelTrace[] trace)
- operatorSummary(Bool currentPeriod)
- run(Natural generations, StoppingRule[] rules, Natural tuningInterval, String checkpointFile, Natural checkpointInterval, Bool underPrior)
Edit this page