com.numericalmethod.suanshu.stats.hmm
Class HmmSim

java.lang.Object
  com.numericalmethod.suanshu.stats.hmm.HmmSim

public class HmmSim
extends java.lang.Object

This class simulates a (discrete) hidden Markov model as defined by Rabiner (1989).

See Also:

"Rabiner, L.R. "A tutorial on hidden Markov models and selected applications in speech recognition," Proceedings of the IEEE. Volume: 77, Issue:2, 257 - 286. Feb 1989."

Constructor Summary

HmmSim(HiddenMarkovModel model)
Construct a simulator for a (discrete) hidden Markov model.

Method Summary

int[]	simObservations(int[] qt) Simulate the observations {O_t} (t = 1, 2, ..., T) for a (discrete) hidden Markov model.
int[]	simStates(int T) Simulate the hidden states {q_t} (t = 1, 2, ..., T) for a (discrete) hidden Markov model.

Methods inherited from class java.lang.Object

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

Constructor Detail

HmmSim

public HmmSim(HiddenMarkovModel model)

Construct a simulator for a (discrete) hidden Markov model.

Parameters:

model - a (discrete) hidden Markov model

Method Detail

simStates

public int[] simStates(int T)

Simulate the hidden states {q_t} (t = 1, 2, ..., T) for a (discrete) hidden Markov model.

Parameters:

T - the length of the (univariate) time series (states and observations) to generate

Returns:

states {q_t} (t = 1, 2, ..., T) with the given (discrete) HMM specification.

simObservations

public int[] simObservations(int[] qt)

Simulate the observations {O_t} (t = 1, 2, ..., T) for a (discrete) hidden Markov model.

Parameters:

qt - the hidden states {q_t}

Returns:

observations {O_t} (t = 1, 2, ..., T)