public class LinearKalmanFilter extends Object
Constructor and Description |
---|
LinearKalmanFilter(DLM model)
Construct a Kalman filter from a univariate controlled dynamic linear model.
|
Modifier and Type | Method and Description |
---|---|
void |
filtering(double[] Y)
Filter the observations without control variable.
|
void |
filtering(double[] Y,
double[] U)
Filter the observations.
|
double |
getFittedState(int t)
Get the posterior expected state.
|
double[] |
getFittedStates()
Get the posterior expected states.
|
double |
getFittedStateVariance(int t)
Get the posterior expected state variance.
|
double |
getKalmanGain(int t)
Get the Kalman gain.
|
double |
getPredictedObservation(int t)
Get the prior observation prediction.
|
double[] |
getPredictedObservations()
Get the prior observation predictions.
|
double |
getPredictedObservationVariance(int t)
Get the prior observation prediction variance.
|
double |
getPredictedState(int t)
Get the prior expected state.
|
double[] |
getPredictedStates()
Get the prior expected states.
|
double |
getPredictedStateVariance(int t)
Get the prior expected state variance.
|
int |
size()
Get T, the number of hidden states or observations.
|
public LinearKalmanFilter(DLM model)
model
- a univariate controlled DLMpublic void filtering(double[] Y, double[] U)
Y
- the observationsU
- the controlspublic void filtering(double[] Y)
Y
- the observationspublic int size()
public double[] getFittedStates()
public double[] getPredictedStates()
public double[] getPredictedObservations()
public double getFittedState(int t)
t
- time, t ≥ 1public double getFittedStateVariance(int t)
t
- time, t ≥ 1public double getPredictedState(int t)
t
- time, t ≥ 1public double getPredictedStateVariance(int t)
t
- time, t ≥ 1public double getPredictedObservation(int t)
t
- time, t ≥ 1public double getPredictedObservationVariance(int t)
t
- time, t ≥ 1public double getKalmanGain(int t)
t
- time, t ≥ 1Copyright © 2010-2020 NM FinTech Ltd.. All Rights Reserved.