The Causality Horizon: Limitations to Predictability of Behavior Using Fuzzy Inductive Reasoning

Keywords

Abstract

This paper focuses on limitations to predictability of system behavior. The concept of a causality horizon is introduced that helps determine the likelihood of success of a qualitative prediction. The Fuzzy Inductive Reasoning (FIR) methodology is the tool used in this paper to model and simulate (forecast) the system behavior.

Two systems, a linear state-space model and a biomedical system, serve to demonstrate the concept. These very different types of systems were chosen in order to illustrate the differences in predictability between applications from technical domains and those from soft sciences areas. It is much more difficult to obtain decent predictions for soft science systems than to obtain accurate predictions for systems from the hard sciences. The purpose of this paper is to explain this discrepancy.

Interested in reading the full paper? (5 pages, 382,887 bytes, pdf)

Homepage

Last modified: January 19, 2006 -- © François Cellier