Compatible deep neural network framework with financial time series data, including data preprocessor, neural network model and trading strategy

TL;DR

This paper presents a comprehensive deep learning framework for financial time series prediction, including novel data preprocessing, a specialized neural network architecture, and a trading strategy, validated on multiple datasets for profitability.

Contribution

It introduces a new data preprocessing pipeline with feature engineering and autoencoders, along with a tailored neural network model and trading strategy for financial prediction.

Findings

Framework yields profitable predictions on three datasets

Data preprocessing improves model accuracy

Model demonstrates robustness across different market data

Abstract

Experience has shown that trading in stock and cryptocurrency markets has the potential to be highly profitable. In this light, considerable effort has been recently devoted to investigate how to apply machine learning and deep learning to interpret and predict market behavior. This research introduces a new deep neural network architecture and a novel idea of how to prepare financial data before feeding them to the model. In the data preparation part, the first step is to generate many features using technical indicators and then apply the XGBoost model for feature engineering. Splitting data into three categories and using separate autoencoders, we extract high-level mixed features at the second step. This data preprocessing is introduced to predict price movements. Regarding modeling, different convolutional layers, an long short-term memory unit, and several fully-connected layers have been designed to perform binary classification. This research also introduces a trading strategy to exploit the trained model outputs. Three different datasets are used to evaluate this method, where results indicate that this framework can provide us with profitable and robust predictions.