Package jline.api.mam

Class Map_dist_acfKt

    • Nested Class Summary

      Nested Classes 
      Modifier and Type Class Description

    • Field Summary

      Fields 
      Modifier and Type Field Description

    • Constructor Summary

      Constructors 
      Constructor Description

    • Enum Constant Summary

      Enum Constants 
      Enum Constant Description

    • Constructor Detail

    • Method Detail

      • map_geo_mul_sum

         final static Double map_geo_mul_sum(Matrix A0, Matrix A1, Matrix B0, Matrix B1, Matrix alA, Matrix alB)

        Computes the geometric sum needed for the autocorrelation distance.

        Solves a discrete Lyapunov equation involving the centered embedded chains of the two MAPs.

        Parameters:
        A0 - hidden transition matrix of the first MAP
        A1 - visible transition matrix of the first MAP
        B0 - hidden transition matrix of the second MAP
        B1 - visible transition matrix of the second MAP
        alA - stationary vector at arrivals of the first MAP
        alB - stationary vector at arrivals of the second MAP
        Returns:

        the geometric sum value (scalar)

      • map_dist_acf

         final static Double map_dist_acf(Matrix A0, Matrix A1, Matrix B0, Matrix B1)

        Computes the squared L2 distance between autocorrelation functions of two MAPs.

        The stationary vectors at arrivals are computed internally.

        Parameters:
        A0 - hidden transition matrix of the first MAP
        A1 - visible transition matrix of the first MAP
        B0 - hidden transition matrix of the second MAP
        B1 - visible transition matrix of the second MAP
        Returns:

        the squared L2 distance between ACFs

      • map_dist_acf

         final static Double map_dist_acf(Matrix A0, Matrix A1, Matrix B0, Matrix B1, Matrix alA, Matrix alB)

        Computes the squared L2 distance between autocorrelation functions of two MAPs.

        Uses the formula based on geometric sums and second-order moments, normalized by the variances of the inter-arrival time distributions.

        Parameters:
        A0 - hidden transition matrix of the first MAP
        A1 - visible transition matrix of the first MAP
        B0 - hidden transition matrix of the second MAP
        B1 - visible transition matrix of the second MAP
        alA - stationary vector at arrivals of the first MAP
        alB - stationary vector at arrivals of the second MAP
        Returns:

        the squared L2 distance between ACFs