Global Utilities

Seminars - Abstract

Department of Computer Science & Computer Engineering

Topic:   Lag-Dependent Regularization for MLPs Applied to Financial Time Series Forecasting Tasks
Speaker:   Andrew Skabar
Date:   01-06-2009
Time:   3:00 PM
Venue:   SEMS meeting room, Bundoora
Abstract:   The application of multilayer perceptrons to forecasting the future value of some time series based on past (or lagged) values of the time series usually requires very careful selection of the number of lags to be used as inputs, and this must usually be determined empirically. This paper proposes a regularization technique by which the influence that a lag has in determining the forecast value decreases exponentially with the lag, and is consistent with the intuitive notion that recent values should have more influence than less recent values in predicting future values. This means that in principle an infinite number of lags could be used. Empirical results show that the regularization technique yields superior performance on out-of-sample data compared with approaches that use a fixed number of inputs without lag-dependent regularization.
Content Approved by: Head of School
Page maintained by: Applications Programmer
Last Updated: 14 October, 2009

Seminar Information