Advertisement

Introduction

  • Dirk Husmeier
Part of the Perspectives in Neural Computing book series (PERSPECT.NEURAL)

Abstract

One of the central problems in science is forecasting. If we have some knowledge about the past behaviour of a system, how can we make meaningful predictions about its future? A standard approach adopted in physics is to construct a mathematical model based on theoretical considerations and, using measured data for specifying the initial conditions, try to integrate the equations of motion forward in time to predict the future state. This procedure, however, is not always feasible. In fields such as economics we still lack the first principles necessary to make good models. In other cases, such as fluid flow, initial data are difficult to obtain. Finally, in complex nonlinear systems with many degrees of freedom, like a turbulent fluid, the weather, or the economy, it is not possible to solve the equations of dynamics explicitly and to keep track of motion in the high dimensional state space. In these cases model-based forecasting becomes impossible and calls for a different prediction paradigm.

Keywords

Stochastic Perturbation Embedding Theorem Time Series Prediction Dynamic Noise Observational Noise 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag London Limited 1999

Authors and Affiliations

  • Dirk Husmeier
    • 1
  1. 1.Neural Systems Group, Department of Electrical & Electronic EngineeringImperial CollegeLondonUK

Personalised recommendations