Skip to content

HiddenMarkovModel

class HiddenMarkovModel<Parameter, State, Observation> < MarkovModel<Parameter, State>

Hidden Markov Model (HMM) or state-space model (SSM).

The joint distribution is:

Typically the are observed, although they need not be.

Graphical model depicting HiddenMarkovModel/StateSpaceModel.

A model inheriting from HiddenMarkovModel/StateSpaceModel overrides the parameter, initial, transition and observation member fibers to specify the individual components of the joint distribution.

The Model class hierarchy is as follows:

Member Variables

Name Description
y:Iterator<Observation> Observations.

Member Fibers

Name Description
simulate Step.
observation Observation model.

Member Fiber Details

observation

fiber observation(y:Observation, x:State, θ:Parameter) -> Event

Observation model.

  • y: The observations, to be set.
  • x: The current state.
  • θ: The parameters.

simulate

fiber simulate(t:Integer) -> Event

Step. Simulates the initial state, or the transition to the next state, and an observation.