Package org.hipparchus.stat.fitting

Class MultivariateNormalMixtureExpectationMaximization

  • public class MultivariateNormalMixtureExpectationMaximization
extends Object
    Expectation-Maximization algorithm for fitting the parameters of multivariate normal mixture model distributions. This implementation is pure original code based on EM Demystified: An Expectation-Maximization Tutorial by Yihua Chen and Maya R. Gupta, Department of Electrical Engineering, University of Washington, Seattle, WA 98195. It was verified using external tools like CRAN Mixtools (see the JUnit test cases) but it is not based on Mixtools code at all. The discussion of the origin of this class can be seen in the comments of the MATH-817 JIRA issue.

    • Method Summary

      All Methods Static Methods Instance Methods Concrete Methods 
      Modifier and Type Method Description
      static org.hipparchus.distribution.multivariate.MixtureMultivariateNormalDistribution estimate​(double[][] data, int numComponents)
      Helper method to create a multivariate normal mixture model which can be used to initialize fit(MixtureMultivariateNormalDistribution).
      void fit​(org.hipparchus.distribution.multivariate.MixtureMultivariateNormalDistribution initialMixture)
      Fit a mixture model to the data supplied to the constructor.
      void fit​(org.hipparchus.distribution.multivariate.MixtureMultivariateNormalDistribution initialMixture, int maxIterations, double threshold)
      Fit a mixture model to the data supplied to the constructor.
      org.hipparchus.distribution.multivariate.MixtureMultivariateNormalDistribution getFittedModel()
      Gets the fitted model.
      double getLogLikelihood()
      Gets the log likelihood of the data under the fitted model.

    • Constructor Detail

      • MultivariateNormalMixtureExpectationMaximization

        public MultivariateNormalMixtureExpectationMaximization​(double[][] data)
                                                 throws org.hipparchus.exception.MathIllegalArgumentException
        Creates an object to fit a multivariate normal mixture model to data.
        Parameters:
        data - Data to use in fitting procedure
        Throws:
        org.hipparchus.exception.MathIllegalArgumentException - if data has no rows
        org.hipparchus.exception.MathIllegalArgumentException - if rows of data have different numbers of columns
        org.hipparchus.exception.MathIllegalArgumentException - if the number of columns in the data is less than 2

    • Method Detail

      • fit

        public void fit​(org.hipparchus.distribution.multivariate.MixtureMultivariateNormalDistribution initialMixture,
                int maxIterations,
                double threshold)
         throws org.hipparchus.exception.MathIllegalArgumentException
        Fit a mixture model to the data supplied to the constructor. The quality of the fit depends on the concavity of the data provided to the constructor and the initial mixture provided to this function. If the data has many local optima, multiple runs of the fitting function with different initial mixtures may be required to find the optimal solution. If a MathIllegalArgumentException is encountered, it is possible that another initialization would work.
        Parameters:
        initialMixture - Model containing initial values of weights and multivariate normals
        maxIterations - Maximum iterations allowed for fit
        threshold - Convergence threshold computed as difference in logLikelihoods between successive iterations
        Throws:
        org.hipparchus.exception.MathIllegalArgumentException - if any component's covariance matrix is singular during fitting
        org.hipparchus.exception.MathIllegalArgumentException - if numComponents is less than one or threshold is less than Double.MIN_VALUE
        org.hipparchus.exception.MathIllegalArgumentException - if initialMixture mean vector and data number of columns are not equal

      • fit

        public void fit​(org.hipparchus.distribution.multivariate.MixtureMultivariateNormalDistribution initialMixture)
         throws org.hipparchus.exception.MathIllegalArgumentException
        Fit a mixture model to the data supplied to the constructor. The quality of the fit depends on the concavity of the data provided to the constructor and the initial mixture provided to this function. If the data has many local optima, multiple runs of the fitting function with different initial mixtures may be required to find the optimal solution. If a MathIllegalArgumentException is encountered, it is possible that another initialization would work.
        Parameters:
        initialMixture - Model containing initial values of weights and multivariate normals
        Throws:
        org.hipparchus.exception.MathIllegalArgumentException - if any component's covariance matrix is singular during fitting
        org.hipparchus.exception.MathIllegalArgumentException - if numComponents is less than one or threshold is less than Double.MIN_VALUE

      • estimate

        public static org.hipparchus.distribution.multivariate.MixtureMultivariateNormalDistribution estimate​(double[][] data,
                                                                                                      int numComponents)
                                                                                               throws org.hipparchus.exception.MathIllegalArgumentException
        Helper method to create a multivariate normal mixture model which can be used to initialize fit(MixtureMultivariateNormalDistribution). This method uses the data supplied to the constructor to try to determine a good mixture model at which to start the fit, but it is not guaranteed to supply a model which will find the optimal solution or even converge.
        Parameters:
        data - Data to estimate distribution
        numComponents - Number of components for estimated mixture
        Returns:
        Multivariate normal mixture model estimated from the data
        Throws:
        org.hipparchus.exception.MathIllegalArgumentException - if numComponents is greater than the number of data rows.
        org.hipparchus.exception.MathIllegalArgumentException - if numComponents < 2.
        org.hipparchus.exception.MathIllegalArgumentException - if data has less than 2 rows
        org.hipparchus.exception.MathIllegalArgumentException - if rows of data have different numbers of columns

      • getLogLikelihood

        public double getLogLikelihood()
        Gets the log likelihood of the data under the fitted model.
        Returns:
        Log likelihood of data or zero of no data has been fit

      • getFittedModel

        public org.hipparchus.distribution.multivariate.MixtureMultivariateNormalDistribution getFittedModel()
        Gets the fitted model.
        Returns:
        fitted model or null if no fit has been performed yet.