Interval forecasting using artificial neural networks trained with Monte Carlo Markov Chain methods

Recent work showed that Bayesian formulation of the neural networks' training problem provide a

This study employs Markov Chain Monte Carlo methods, which make possible feasible

time series forecasting. [t exploits the advantages of the 'mean absolute deviations' error function (as

uniformity of the prediction error bounds over the state space or where the assumption of normally

estimators of the 'prediction intervals' associated with a prespecified probability level. The length of

state space as well as in m-step ahead predictions.

model with Gaussian and non-Gaussian disturbances, (b) the logistic map (chaotic) with added

Gaussian noise, and finally, (c) a real-world financial time series.