Predicting Jump Arrivals in Stock Prices Using Neural Networks with Limit Order Book Data

Abstract

This paper proposes a new method for predicting jump arrivals in stock markets with high-frequency limit order book data. We introduce a new model architecture, based on Convolutional Long Short-Term Memory with attention, to apply time series representation learning with memory and to focus the prediction attention on the most important features to improve performance. Using order book data on five liquid U.S. stocks, we provide empirical evidence on the efficacy of the proposed approach. We find that the proposed approach with an attention mechanism outperforms the multi-layer perceptron network as well as the convolutional neural network and Long Short-Term memory model. The use of limit order book data was found to improve the performance of the proposed model in jump prediction, either clearly or marginally, depending on the underlying stock.