Lag-ℓ forecasting and machine-learning algorithms

Abstract

In this study, we discuss the problem of lag-ℓ forecasting, which has been solved using a machine-learning algorithm. The main aim of this study is to define the lag-ℓ forecaster based on a precise classification approach and to discuss the technical issues involved in the lag-ℓ forecasting problem, including a comparison of various machine-learning algorithms for proper implementation of the technique. This study focuses on an application that uses the lag-ℓ forecaster in an early-warning system. © 2011 Blackwell Publishing Ltd.