smile.regression

Class RLS

java.lang.Object

smile.regression.RLS

All Implemented Interfaces:

java.io.Serializable, OnlineRegression<double[]>, Regression<double[]>

public class RLS
extends java.lang.Object
implements OnlineRegression<double[]>, java.io.Serializable

Recursive least squares. RLS updates an ordinary least squares with samples that arrive sequentially. To initialize RLS, we typically train an OLS model with a batch of samples. In some adaptive configurations it can be useful not to give equal importance to all the historical data but to assign higher weights to the most recent data (and then to forget the oldest one). This may happen when the phenomenon underlying the data is non stationary or when we want to approximate a nonlinear dependence by using a linear model which is local in time. Both these situations are common in adaptive control problems.

References

1. https://www.otexts.org/1582

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	RLS.Trainer Trainer for linear regression by recursive least squares.

Constructor Summary

Constructors

Constructor and Description
RLS(double[][] x, double[] y) Constructor.
RLS(double[][] x, double[] y, double lambda) Constructor.

Method Summary

Modifier and Type	Method and Description
double[]	coefficients() Returns the linear coefficients, of which the last element is the intercept.
double	getForgettingFactor() Get the forgetting factor
void	learn(double[][] x, double[] y) Learn a new instance with online regression.
void	learn(double[] x, double y) Learn a new instance with online regression.
double	predict(double[] x) Predicts the dependent variable of an instance.
void	setForgettingFactor(double lambda) Set the forgetting factor

Methods inherited from class java.lang.Object

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

Methods inherited from interface smile.regression.Regression

predict

Constructor Detail

RLS

public RLS(double[][] x,
           double[] y)

Constructor. Learn the ordinary least squares model to initialize gamma and coefficients.

Parameters:

x - a matrix containing the explanatory variables. NO NEED to include a constant column of 1s for bias.

y - the response values.

RLS

public RLS(double[][] x,
           double[] y,
           double lambda)

Constructor. Learn the ordinary least squares model to initialize gamma and coefficients.

Parameters:

x - a matrix containing the explanatory variables. NO NEED to include a constant column of 1s for bias.

y - the response values.

lambda - the forgetting factor.

Method Detail

coefficients

public double[] coefficients()

Returns the linear coefficients, of which the last element is the intercept.

predict

public double predict(double[] x)

Description copied from interface: Regression

Predicts the dependent variable of an instance.

Specified by:

predict in interface Regression<double[]>

Parameters:

x - the instance.

Returns:

the predicted value of dependent variable.

learn

public void learn(double[][] x,
                  double[] y)

Learn a new instance with online regression.

Parameters:

x - the training instances.

y - the target values.

learn

public void learn(double[] x,
                  double y)

Learn a new instance with online regression.

Specified by:

learn in interface OnlineRegression<double[]>

Parameters:

x - the training instance.

y - the target value.

getForgettingFactor

public double getForgettingFactor()

Get the forgetting factor

Returns:

the forgetting factor

setForgettingFactor

public void setForgettingFactor(double lambda)

Set the forgetting factor

Parameters:

lambda - the forgetting factor