Fractional Gaussian Noise

Initializing live version
Download to Desktop

Requires a Wolfram Notebook System

Interact on desktop, mobile and cloud with the free Wolfram Player or other Wolfram Language products.

This Demonstration generates a fractional Gaussian noise time series of length with Hurst coefficient , mean 0, and variance 1. The plot shows 100 successive observations at a time. To avoid clutter, only the center is labeled on the horizontal bottom frame. By sliding the "start" bar, you can view successive subsequences.

[more]

When , the series exhibits long-range dependence, whose strength increases with . Series with typically have long runs above and below the mean.

For , the series is dominated by high-frequency noise and is sometimes called anti-persistent.

The case corresponds to Gaussian white noise.

To visualize the entire time series, click the "plus" button to the right of the "start" slider to show the animation menu, then click the "play" button.

By using the "random seed" slider, you can view other time series.

[less]

Contributed by: Ian McLeod (January 2012)
(University of Western Ontario)
Open content licensed under CC BY-NC-SA


Snapshots


Details

[1] provides a review of the fractional Gaussian noise and related long-memory time series models.

Reference

[1] K. W. Hipel and A. I. McLeod, Time Series Modelling of Water Resources and Environmental Systems, Amsterdam: Elsevier, 1994.



Feedback (field required)
Email (field required) Name
Occupation Organization
Note: Your message & contact information may be shared with the author of any specific Demonstration for which you give feedback.
Send