Mcmc.Status

Creation date: Tue May 5 18:01:15 2020.

Documentation

data Cleaner a

Clean the state periodically.

The prior and the likelihood will be updated after the cleaning process.

For long chains, successive numerical errors can accumulate such that the state diverges from honoring specific required constraints. In these cases, a Cleaner can be used to ensure that the required constraints of the state are honored. For example, the branches of an ultrametric phylogeny may diverge slightly after successful many proposals such that the phylogeny is not anymore ultrametric.

Please be aware that the Markov chain will not converge to the true posterior distribution if the state is changed substantially! Only apply subtle changes that are absolutely necessary to preserve the required properties of the state such as specific numerical constraints.

Constructors

Cleaner

Fields

• clEvery :: Int

  Clean every given number of iterations.

• clFunction :: a -> a

  Cleaning function. Executed before monitoring the state.

data Status a

The Status contains all information to run an MCMC chain. It is constructed using the function status.

The polymorphic type a stores the state of the chain. It can also be used to store auxiliary information.

Constructors

Status

Fields

• name :: String

  The name of the MCMC chain; used as file prefix.

• item :: Item a

  The current Item of the chain combines the current state and the current likelihood.

• iteration :: Int

  The iteration is the number of completed cycles.

• trace :: Trace a

  The Trace of the Markov chain in reverse order, the most recent Item is at the head of the list.

• acceptance :: Acceptance (Proposal a)

  For each Proposal, store the list of accepted (True) and rejected (False) proposals; for reasons of efficiency, the list is also stored in reverse order.

• burnInIterations :: Maybe Int

  Number of burn in iterations; deactivate burn in with Nothing.

• autoTuningPeriod :: Maybe Int

  Auto tuning period (only during burn in); deactivate auto tuning with Nothing.

• iterations :: Int

  Number of normal iterations excluding burn in. Note that auto tuning only happens during burn in.

• forceOverwrite :: Bool

  Overwrite output files? Default is False, change with force.

• save :: Maybe Int

  Save the chain with trace of given length at the end of the run? Default is no save (Nothing). Change with saveWith.

• verbosity :: Verbosity

  Verbosity.

• generator :: GenIO

  The random number generator.

• start :: Maybe (Int, UTCTime)

  Starting time and starting iteration of chain; used to calculate run time and ETA.

• logHandle :: Maybe Handle

  Handle to log file.

• priorF :: a -> Log Double

  The prior function. The un-normalized posterior is the product of the prior and the likelihood.

• likelihoodF :: a -> Log Double

  The likelihood function. The un-normalized posterior is the product of the prior and the likelihood.

• cleaner :: Maybe (Cleaner a)

  Clean the state periodically.

• cycle :: Cycle a

  A set of Proposals form a Cycle.

• monitor :: Monitor a

  A Monitor observing the chain.

status

Arguments

:: String

Name of the Markov chain; used as file prefix.

-> (a -> Log Double)

The prior function.

-> (a -> Log Double)

The likelihood function.

-> Cycle a

A list of Proposals executed in forward order. The chain will be logged after each cycle.

-> Monitor a

A Monitor observing the chain.

-> a

The initial state in the state space a.

-> Maybe Int

Number of burn in iterations; deactivate burn in with Nothing.

-> Maybe Int

Auto tuning period (only during burn in); deactivate auto tuning with Nothing.

-> Int

Number of normal iterations excluding burn in. Note that auto tuning only happens during burn in.

-> GenIO

A source of randomness. For reproducible runs, make sure to use generators with the same, fixed seed.

-> Status a

Initialize the Status of a Markov chain Monte Carlo run.

cleanWith :: Cleaner a -> Status a -> Status a

Clean the state every given number of generations using the given function. See Cleaner.

saveWith :: Int -> Status a -> Status a

Save the Markov chain with trace of given length.

force :: Status a -> Status a

Overwrite existing files; it is not necessary to use force, when a chain is continued.

quiet :: Status a -> Status a

Do not print anything to standard output. Do not create log file. File monitors and batch monitors are executed normally.

debug :: Status a -> Status a

Be verbose.

noData :: Status a -> Status a

Set the likelihood function to 1.0. Useful for debugging and testing.