Class MultivariateAutoCorrelationFunction

  • All Implemented Interfaces:
    Function<Vector,​Matrix>, RntoMatrix
    Direct Known Subclasses:
    VARMAAutoCorrelation

    public abstract class MultivariateAutoCorrelationFunction
    extends R2toMatrix
    This is the auto-correlation function of a multi-dimensional time series {Xt}. \[ Y_{i,j} = E((X_i - \mu_i) \times (X_j - \mu_j)') \\ \rho(i, j) = \frac{Y_{i,j}}{\sqrt{Y_{i,i}Y_{j,j}}} \] For a stationary process, the auto-correlation depends only on the lag, |i - j|.
    • Constructor Detail

      • MultivariateAutoCorrelationFunction

        public MultivariateAutoCorrelationFunction()
    • Method Detail

      • get

        public Matrix get​(int i,
                          int j)
        Get the auto-correlation of Xi and Xj.
        Parameters:
        i - i > 0
        j - j > 0
        Returns:
        the auto-correlation Matrix indexed by [i, j]