smile.stat.distribution

Interface MultivariateExponentialFamily

All Known Implementing Classes:

MultivariateGaussianDistribution

public interface MultivariateExponentialFamily

The purpose of this interface is mainly to define the method M that is the Maximization step in the EM algorithm. Note that distributuions of exponential family has the close-form solutions in the EM algorithm. With this interface, we may allow the mixture contains distributions of different form as long as it is from exponential family.

Author:

Haifeng Li

See Also:

ExponentialFamilyMixture

Method Summary

Methods

Modifier and Type	Method and Description
MultivariateMixture.Component	M(double[][] x, double[] posteriori) The M step in the EM algorithm, which depends the specific distribution.

Method Detail

M

MultivariateMixture.Component M(double[][] x,
                              double[] posteriori)

The M step in the EM algorithm, which depends the specific distribution.

Parameters:

x - the input data for estimation

posteriori - the posteriori probability.

Returns:

the (unnormalized) weight of this distribution in the mixture.