Details of Grant 

EPSRC Reference:	GR/K51815/01
Title:	NONSTATIONARY FEATURE EXTRACTION & TRACKING FOR THE CLASSIFICATION OF TURNING POINTS IN MULTIVARIATE TIME
Principal Investigator:	Lowe, Professor D
Other Investigators:
Researcher Co-Investigators:
Project Partners:	Pre Nexus Migration
Department:	Electronic Engineering
Organisation:	Aston University
Scheme:	Standard Research (Pre-FEC)
Starts:	13 May 1996	Ends:	12 May 1999	Value (£):	139,456
EPSRC Research Topic Classifications:
EPSRC Industrial Sector Classifications:	Financial Services Energy
Related Grants:
Panel History:
Summary on Grant Application Form
The project will develop techniques, theory and algorithms, for the extraction and tracking of 'structure' or features in generally multivariate nonstationary time series based on artificial neural networks methods. Almost all recent advances in artificial neural network research are based on the principle of 'stationary'; it is implicitly assumed that the underlying generator of the observed data is not itself time varying. Unfortunately many problems which occur in the real world are explicitly nonstationary. The proposed research programme driven by real world applications, will address some of the limitations of current network methods by developing approaches to extract, characterise and track structural features from nonstationary, non-linear time series. Data and background will be provided by financial and industrial collaborators. The focus of the research is towards feature extraction, or structure characterisation, using unsupervised and 'relative supervision' networks based on the underpinning philosophy of Information Geometry. Specific approaches will be developed based upon Nonlinear Principal Component Subspaces (related to the method of Principal Xcurves) and unconditional Mixture Density networks. Although these neural network techniques in themselves do not deal with non-stationarity, the fact that they are closely parameterised mappings means that optimisation techniques can be adapted to track nonstationarity provided it may be characterised on a slow enough time scale.
Key Findings
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Potential use in non-academic contexts
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Impacts
Description This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk Summary Date Materialised
Sectors submitted by the Researcher
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Project URL:
Further Information:
Organisation Website:	http://www.aston.ac.uk