org.omegahat.Simulation.MCMC.Proposals
Class AdaptiveNormalMetropolisProposal
java.lang.Object
|
+--org.omegahat.Simulation.MCMC.Proposals.AdaptiveProposal
|
+--org.omegahat.Simulation.MCMC.Proposals.AdaptiveNormalMetropolisProposal
- All Implemented Interfaces:
- HastingsCoupledProposal
- public class AdaptiveNormalMetropolisProposal
- extends AdaptiveProposal
Methods inherited from class java.lang.Object |
, clone, equals, finalize, getClass, hashCode, notify, notifyAll, registerNatives, toString, wait, wait, wait |
inflationFactor
protected double inflationFactor
DEBUG
protected boolean DEBUG
minVar
protected double[][] minVar
maxVar
protected double[][] maxVar
AdaptiveNormalMetropolisProposal
public AdaptiveNormalMetropolisProposal(double[][] var,
PRNG prng)
- Constructor for normal increment proposal with specified
covariance matrix, no variance inflation @param var variance matrix
AdaptiveNormalMetropolisProposal
public AdaptiveNormalMetropolisProposal(double[][] var,
double inflationFactor,
PRNG prng)
- Constructor for normal increments with specified covariance matrix.
- Parameters:
var
- variance matrix
AdaptiveNormalMetropolisProposal
public AdaptiveNormalMetropolisProposal(int length,
double var,
PRNG prng)
- Constructor for a independent normal increntens (sam variances for each dimension).
No variance inflation.
The diagonal elements of the covariance matrix will be set to the specified value, with off diagonals set to 0
- Parameters:
length
- number of dimensionsvar
- diagnonal values for covariance matrix (off diagonals are set to 0)
AdaptiveNormalMetropolisProposal
public AdaptiveNormalMetropolisProposal(int length,
double var,
double inflationFactor,
PRNG prng)
- Constructor for a spherical normal increments with the same variance for each dimension.
The diagonal elements of the covariance matrix will be set to the specified value, with off diagonals set to 0
- Parameters:
length
- number of dimensionsvar
- diagnonal values for covariance matrix (off diagonals are set to 0)inflationFactor
- factor to inflate observed variance when adapting.
getInflationFactor
public double getInflationFactor()
setInflationFactor
public double setInflationFactor(double in)
setMinVar
public double[][] setMinVar(double[][] var)
setMaxVar
public double[][] setMaxVar(double[][] var)
adapt
public void adapt(MultiState mstate)
- Description copied from class:
AdaptiveProposal
- modify the state of the enclosed proposal distribution using information from
the provided state vector
- Overrides:
adapt
in class AdaptiveProposal
main
public static void main(java.lang.String[] argv)