ffGn: Fast Fractional Gaussian Noise

In wol-fi/multifractal: Multifractal and Surrogate Analysis

Fast Fractional Gaussian Noise

Description

A fast algorithm for simulating fractal Gaussian noise.

Usage

ffGn(n=1000, H=0.5, mu=0, sigma=1)

Arguments

n	length of the simulated series.
H	a numeric value within 0 and 1 denoting the Hurst exponent. This describes the persistence (i.e., level of linear auto-correlation). Above 0.5 is persistent, below 0.5 is anti-persistent.
mu	mean value of the series.
sigma	standard deviation of the series.

Value

Returns a series of simulated fractal Gaussian noise.

Note

Translated from Matlab into R. Original Matlab code by B. Scott Jackson.

Author(s)

Wolfgang Schadner

References

Mandelbrot, B. B., Fisher, A. J., & Calvet, L. E. (1997). A multifractal model of asset returns.

See Also

mfsim, mfdfa

Examples

  x <- ffGn(H=0.8)
  
  # fractal Brownian motion:
  B <- cumsum(x)
  plot(B)

wol-fi/multifractal documentation built on May 31, 2022, 1:18 a.m.