Title :
Neural expert weighing
Author :
de O Valle dos Santos, Rafael ; Vellasco, Marley M B R
Author_Institution :
Electr. Eng. Dept., Pontifical Catholic Univ. of Rio de Janeiro (PUC-Rio), Rio de Janeiro, Brazil
Abstract :
This article describes a novel framework for combining time series forecasts. It uses neural network regression models to estimate, at a given point in time, the linear weights (relevancies) of the available experts (forecasters) at that time. With those weights, the experts can be linearly combined to produce a single, potentially more accurate, forecast. This new weight generation framework was designed to be especially useful for multi-step-ahead forecasting.
Keywords :
neural nets; regression analysis; time series; linear weights; multistep-ahead forecasting; neural expert weighing; neural network regression models; time series forecasts; Adaptation model; Artificial neural networks; Forecasting; Least squares approximation; Predictive models; Time series analysis; Training;
Conference_Titel :
Neural Networks (IJCNN), The 2010 International Joint Conference on
Conference_Location :
Barcelona
Print_ISBN :
978-1-4244-6916-1
DOI :
10.1109/IJCNN.2010.5596879
