Design of High-Frequency Trading Algorithm Based on Machine Learning

TL;DR

This paper introduces a machine learning-based framework for high-frequency trading that integrates deep learning optimization techniques with market data analysis to improve liquidity prediction and trading performance.

Contribution

It develops a novel analytical framework combining VPIN, GARCH, and SVM to better utilize order book data for high-frequency trading strategies.

Findings

VPIN effectively predicts market liquidity.

Model backtested on CSI300 futures shows improved return predictions.

Integration of deep learning enhances trading decision accuracy.

Abstract

Based on iterative optimization and activation function in deep learning, we proposed a new analytical framework of high-frequency trading information, that reduced structural loss in the assembly of Volume-synchronized probability of Informed Trading ($VPIN$), Generalized Autoregressive Conditional Heteroscedasticity (GARCH) and Support Vector Machine (SVM) to make full use of the order book information. Amongst the return acquisition procedure in market-making transactions, uncovering the relationship between discrete dimensional data from the projection of high-dimensional time-series would significantly improve the model effect. $VPIN$ would prejudge market liquidity, and this effectiveness backtested with CSI300 futures return.