com.numericalmethod.suanshu.stats.hmm

Class HiddenMarkovModel

java.lang.Object

com.numericalmethod.suanshu.stats.markovchain.SimpleMC

com.numericalmethod.suanshu.stats.hmm.HiddenMarkovModel

All Implemented Interfaces:

RandomNumberGenerator, java.io.Serializable

Direct Known Subclasses:

HiddenMarkovModel, HiddenMarkovModel

public class HiddenMarkovModel
extends SimpleMC

In a (discrete) hidden Markov model, the state is not directly visible, but output, dependent on the state, is visible. Each state has a probability distribution over the possible output tokens (could be continuous). Therefore the sequence of tokens generated by an HMM gives some information about the sequence of states. Note that the adjective 'hidden' refers to the state sequence through which the model passes, not to the parameters of the model; even if the model parameters are known exactly, the model is still 'hidden'. In other words, a hidden Markov model is a Markov chain of (hidden) states and for each state a conditional random number generator (distribution).

See Also:

Wikipedia: Hidden Markov model, Serialized Form

Constructor Summary

Constructors

Constructor and Description
HiddenMarkovModel(HiddenMarkovModel that) Copy constructor.
HiddenMarkovModel(Matrix A, RandomNumberGenerator[] B) Construct a hidden Markov model using the stationary probabilities of the initial states.
HiddenMarkovModel(Vector PI, Matrix A, RandomNumberGenerator[] B) Construct a hidden Markov model.

Method Summary

Modifier and Type	Method and Description
HmmInnovation	next() Get the next simulated innovation - state and observation.
double	nextDouble() Get the next simulated observation.
void	seed(long... seeds) Seed the random number generator to produce repeatable sequences.

Methods inherited from class com.numericalmethod.suanshu.stats.markovchain.SimpleMC

A, bin, getStationaryProbabilities, nextState, nStates, PI

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

HiddenMarkovModel

public HiddenMarkovModel(Vector PI,
                         Matrix A,
                         RandomNumberGenerator[] B)

Construct a hidden Markov model.

Parameters:

PI - the initial state probabilities

A - the state transition probabilities of the homogeneous hidden Markov chain

B - the conditional observation random number generators (distributions)

HiddenMarkovModel

public HiddenMarkovModel(Matrix A,
                         RandomNumberGenerator[] B)

Construct a hidden Markov model using the stationary probabilities of the initial states.

Parameters:

A - the state transition probabilities of the homogeneous hidden Markov chain

B - the conditional observation random number generators (distributions)

HiddenMarkovModel

public HiddenMarkovModel(HiddenMarkovModel that)

Copy constructor.

Parameters:

that - a HiddenMarkovModel

Method Detail

seed

public void seed(long... seeds)

Description copied from interface: RandomNumberGenerator

Seed the random number generator to produce repeatable sequences.

Specified by:

seed in interface RandomNumberGenerator

Overrides:

seed in class SimpleMC

Parameters:

seeds - the seeds

nextDouble

public double nextDouble()

Get the next simulated observation.

Specified by:

nextDouble in interface RandomNumberGenerator

Overrides:

nextDouble in class SimpleMC

Returns:

next observation

next

public HmmInnovation next()

Get the next simulated innovation - state and observation.

Returns:

the next HMM innovation