Comparing neural networks with other predictive models in artificial stock market

Abstrakt

A new way of comparing models for forecasting was created. The confronted models are neural networks (feed-forward neural networks and Elman's simple recurrent neural networks), ARMA models (AR and ARMA), random forecast, a trivial forecast of future value by the last known value and moving average forecast. winning model is the one which earns the most money.

Klíčová slova

• Neural networks
• ARMA
• arti cial market