351st Post, Chaos Theory for Forecasting

This is the 351st post to the blog - an important milestone for me

Can the principles of chaos theory help in our forecasting techniques?

I found a very interesting paper from the book,

Managing Organizational Complexity: Philosophy, Theory, and Application
A Volume in: Managing the Complex, pages 167-182.
2005 by Information Age Publishing, Inc.
ISBN: 1-59311-319-6 (cloth), 1-59311-318-8 (paper)

The paper is

CHAOS-BASED PRINCIPLES OF FORECASTING
Liu Hong and Kurt A. Richardson

They explain the forecasting as

Quantitative
Trajectory Predicting
Range Forecasting
Qualitative Futuring

The import of the paper for me is that before getting into any forecast one need to understand how much of system is composed of what are called strange attractors or for that matter other attractors.

Given our obsession with forecasting - technological, business and what not, it is important to understand what are we forecasting and what really can be the limits to accuracy of forecasts!

Lorenz - chaos theory pioneer

Tribute is due to Edward Lorenz whose death marks the end of Chaos theory 1.0. He started the Butterfly Effect story that has become a folklore now.

Small variation in the initial conditions can create entirely different results in any system if same process is followed. This is the essence of Lorenz curves on weather forecasting.

This is really so simple, yet our Six Sigma friends, especially those who are trying to make Software development six sigma compliant really cant understand.

It will be interesting to see how much predictability can be there in software development processes. I have a lurking intuition it is dynamical and chaotic .

Bye bye Lorenz! May God Bless your Soul!