Improving risk-adjusted performance profile of intraday trading models with Neuro-Fuzzy techniques and moving average high-frequency price signals

Abstract

We present a performance comparison of risk-adjusted intraday trading strategies based on dynamic non-linear models using the more traditional Artificial Neural Network, as well as Adaptive Neuro-Fuzzy Systems (ANFIS) and Dynamic Evolving Neuro Fuzzy Systems (DENFIS). The model selection process takes into account the risk-return measures together with flexible position holding periods and a return band filter, employing a dynamic combination of moving average signals. Our results show that these models can be successfully applied to support intraday trading strategies, especially when considering constraints such as transaction costs and trading hours, which existing approaches in the literature do not account for.

26 Figures and Tables

Cite this paper

@inproceedings{Vella2013ImprovingRP, title={Improving risk-adjusted performance profile of intraday trading models with Neuro-Fuzzy techniques and moving average high-frequency price signals}, author={Vince Vella and Wing Lon Ng and Wing Lon Nga}, year={2013} }