Parameter estimation based on stacked regression and evolutionary algorithms

作     者：Hong, X Billings, SA 

作者机构：Univ Sheffield Dept Automat Control & Syst Engn Sheffield S1 3JD S Yorkshire England 

出 版 物：《IEE PROCEEDINGS-CONTROL THEORY AND APPLICATIONS》 (IEE Proc Control Theory Appl)

年 卷 期：1999年第146卷第5期

页      面：406-414页

核心收录：

主　　题：stacked regression MDE evolutionary algorithms statistical analysis Simulation, modelling and identification Optimisation techniques parameter estimation sunspot time series linear-in-the-parameter models Canadian lynx time series OLS algorithm evolutionary computation minimisation mean dispersion error Interpolation and function approximation (numerical analysis) cross-validation parsimony concise model structure PRESS prediction errors Other topics in statistics cross-validated prediction error minimisation least squares approximations forward orthogonal least-squares algorithm 

摘      要：A new parameter-estimation algorithm, which minimises the cross-validated prediction error for linear-in-the-parameter models, is proposed, based on stacked regression and an evolutionary algorithm. It is initially shown that cross-validation is very important for prediction in linear-in-the-parameter models using a criterion called the mean dispersion error (MDE). Stacked regression, which can be regarded as a sophisticated type of cross-validation, is then introduced based on an evolutionary algorithm, to produce a new parameter-estimation algorithm, which preserves the parsimony of a concise model structure that is determined using the forward orthogonal least-squares (OLS) algorithm. The PRESS prediction errors ale used for cross-validation, and the sunspot and Canadian lynx time series are used to demonstrate the new algorithms.