Combination of Biased Artificial Neural Network Forecasters

Artificial neural networks (ANN) have been paramount for modeling and forecasting time series phenomena. In this way it has been usual to suppose that each ANN model generates a white noise as prediction error. However, mostly because of disturbances not captured by each model, it is yet possible that such supposition is violated. On the other hand, to adopt a single ANN model may lead to statistical bias and underestimation of uncertainty. The present paper introduces a two-step maximum likelihood method for correcting and combining ANN models. Applications involving single ANN models for Dow Jones Industrial Average Index and S&P500 series illustrate the usefulness of the proposed framework.

[1]  Nicholas T. Longford.  Inference with the Lognormal Distribution , 2008 .

[2]  Elvezio Ronchetti,et al.  Stock and bond return predictability: the discrimination power of model selection criteria , 2006, Comput. Stat. Data Anal..

[3]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[4]  Renate Sitte,et al.  Neural Networks Approach to the Random Walk Dilemma of Financial Time Series , 2002, Applied Intelligence.

[5]  Kenneth F. Wallis,et al.  Combining forecasts – forty years later , 2011 .

[6]  S. P. Neuman,et al.  Maximum likelihood Bayesian averaging of uncertain model predictions , 2003 .

[7]  Young-Oh Kim,et al.  Combining single-value streamflow forecasts - a review and guidelines for selecting techniques. , 2009 .

[8]  R. Clemen.  Combining forecasts: A review and annotated bibliography , 1989 .

[9]  Desheng Dash Wu,et al.  Short-term load forecasting using bayesian neural networks learned by Hybrid Monte Carlo algorithm , 2012, Appl. Soft Comput..

[10]  Tiago Alessandro Espínola Ferreira,et al.  A New Intelligent System Methodology for Time Series Forecasting with Artificial Neural Networks , 2008, Neural Processing Letters.

[11]  A. Timmermann,et al.  Combining expert forecasts: Can anything beat the simple average? , 2013 .

[12]  Kin Keung Lai,et al.  A novel nonlinear ensemble forecasting model incorporating GLAR and ANN for foreign exchange rates , 2005, Comput. Oper. Res..

[13]  Dimitri P. Solomatine,et al.  Knowledge-based modularization and global optimization of artificial neural network models in hydrological forecasting , 2007, Neural Networks.

[14]  Paulo S. G. de Mattos Neto,et al.  Time series forecasting using a perturbative intelligent system , 2010, GECCO '10.

[15]  A. Timmermann.  Chapter 4 Forecast Combinations , 2006 .

[16]  Kung-Sik Chan,et al.  Time Series Analysis: With Applications in R , 2010 .

[17]  Giuseppe Storti,et al.  A GMM procedure for combining volatility forecasts , 2008, Comput. Stat. Data Anal..

[18]  Luis Ricardo Lage Rodrigues,et al.  Multi-model calibration and combination of tropical seasonal sea surface temperature forecasts , 2014, Climate Dynamics.

[19]  Michael Y. Hu,et al.  Forecasting with artificial neural networks: The state of the art , 1997 .