Off-the-Shelf Neural Network Architectures for Forex Time Series Prediction come at a Cost

TL;DR

This paper compares LSTM and specialized ANN architectures for forex prediction, showing that the specialized model offers better accuracy with fewer resources and faster execution, informing architecture choice in time series forecasting.

Contribution

It provides a comparative analysis demonstrating that a specialized ANN outperforms LSTM architectures in forex prediction regarding accuracy and resource efficiency.

Findings

Specialized ANN achieves higher prediction accuracy.

The specialized model uses less memory and computational power.

Faster execution times are observed with the specialized architecture.

Abstract

Our study focuses on comparing the performance and resource requirements between different Long Short-Term Memory (LSTM) neural network architectures and an ANN specialized architecture for forex market prediction. We analyze the execution time of the models as well as the resources consumed, such as memory and computational power. Our aim is to demonstrate that the specialized architecture not only achieves better results in forex market prediction but also executes using fewer resources and in a shorter time frame compared to LSTM architectures. This comparative analysis will provide significant insights into the suitability of these two types of architectures for time series prediction in the forex market environment.