Forecasting jump arrivals in stock prices: new attention-based network architecture using limit order book data
From MaRDI portal
Publication:5120733
Abstract: The existing literature provides evidence that limit order book data can be used to predict short-term price movements in stock markets. This paper proposes a new neural network architecture for predicting return jump arrivals in equity markets with high-frequency limit order book data. This new architecture, based on Convolutional Long Short-Term Memory with Attention, is introduced to apply time series representation learning with memory and to focus the prediction attention on the most important features to improve performance. The data set consists of order book data on five liquid U.S. stocks. The use of the attention mechanism makes it possible to analyze the importance of the inclusion limit order book data and other input variables. By using this mechanism, we provide evidence that 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. This suggests that path-dependence in limit order book markets is a stock specific feature. Moreover, 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.
Recommendations
- Encoding of high-frequency order information and prediction of short-term stock price by deep learning
- Modelling high-frequency limit order book dynamics with support vector machines
- Deep learning for limit order books
- Using full limit order book for price jump prediction
- Short-term stock price prediction based on limit order book dynamics
Cites work
- A closed-form solution for options with stochastic volatility with applications to bond and currency options
- A stochastic model for order book dynamics
- Axiomatic derivation of the principle of maximum entropy and the principle of minimum cross-entropy
- Can properly discounted projects follow geometric Brownian motion?
- Financial Modelling with Jump Processes
- Jump and volatility dynamics for the S\&P 500: evidence for infinite-activity jumps with non-affine volatility dynamics from stock and option markets
- Learning, information processing and order submission in limit order markets
- Modelling high-frequency limit order book dynamics with support vector machines
- No-arbitrage semi-martingale restrictions for continuous-time volatility models subject to leverage effects, jumps and i.i.d. noise: theory and testable distributional implications
- Noisy time series prediction using recurrent neural networks and grammatical inference
- Statistical Methods for Rates and Proportions
- Statistical pattern recognition
- Stochastic Volatility With an Ornstein–Uhlenbeck Process: An Extension
- Supervised sequence labelling with recurrent neural networks.
- Testing whether jumps have finite or infinite activity
- The Measurement of Observer Agreement for Categorical Data
- Threshold bipower variation and the impact of jumps on volatility forecasting
- Universal features of price formation in financial markets: perspectives from deep learning
- What drives the sensitivity of limit order books to company announcement arrivals?
- Why is equity order flow so persistent?
Cited in
(10)- Forecasting financial time series with Boltzmann entropy through neural networks
- A deep learning approach to estimating fill probabilities in a limit order book
- Using full limit order book for price jump prediction
- Stock market trend prediction using a functional time series approach
- Encoding of high-frequency order information and prediction of short-term stock price by deep learning
- Exploring the attention mechanism in LSTM-based Hong Kong stock price movement prediction
- Deep order flow imbalance: Extracting alpha at multiple horizons from the limit order book
- A machine learning approach to classification for traders in financial markets
- Algorithmic and high-frequency trading problems for Semi-Markov and Hawkes jump-diffusion models
- Attention-based reading, highlighting, and forecasting of the limit order book
This page was built for publication: Forecasting jump arrivals in stock prices: new attention-based network architecture using limit order book data
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5120733)
