Tensor representation in high-frequency financial data for price change prediction

By Dat Thanh Tran, Martin Magris, Juho Kanniainen, Moncef Gabbouj, Alexandros Iosifidis. 2017

63 citations recorded by the source corpus.

Nowadays, with the availability of massive amount of trade data collected, the dynamics of the financial markets pose both a challenge and an opportunity for high frequency traders. In order to take advantage of the rapid, subtle movement of assets in High Frequency Trading (HFT), an automatic algorithm to analyze and detect patterns of price change based on transaction records must be available. The multichannel, time-series representation of financial data naturally suggests tensor-based learning algorithms. In this work, we investigate the effectiveness of two multilinear methods for the mid-price prediction problem against other existing methods. The experiments in a large scale dataset which contains more than 4 millions limit orders show that by utilizing tensor representation, multilinear models outperform vector-based approaches and other competing ones.

Subfields: Computational Mathematics, Management Science and Operations Research, Signal Processing, Time-Series & Forecasting

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mathematics

1. A. Ozbayoglu
2. Omer Berat Sezer
3. M. Ugur Gudelek

In Management Science and Operations Research

1. M. Ugur Gudelek
2. Omer Berat Sezer
3. A. Ozbayoglu
4. Stefan Zohren
5. Weiwei Jiang
6. Stephen J. Roberts
7. Zihao Zhang
8. Longbing Cao

In Signal Processing

1. Stefan Zohren
2. Stephen J. Roberts
3. A. Ozbayoglu
4. Omer Berat Sezer
5. M. Ugur Gudelek

In Time-Series & Forecasting

1. M. Ugur Gudelek
2. Omer Berat Sezer
3. A. Ozbayoglu
4. Stefan Zohren
5. Stephen J. Roberts
6. Zihao Zhang
7. Weiwei Jiang
8. Longbing Cao

Citations in the corpus

• Financial Time Series Forecasting with Deep Learning : A Systematic Literature Review: 2005-2019 (2019, methodology): “[112] The LOB of 5 stocks of Finnish Stock Market 2010 FI-2010 dataset: bid/ask and volume * WMTR, MDA Accuracy, Precision, Recall, F1Score -”
• Deep Learning for Financial Applications : A Survey (2020, methodology): “[39] developed a DL model that implements price change forecasting through midprice prediction using high-frequency limit order book data with tensor representation.”
• DeepLOB: Deep Convolutional Neural Networks for Limit Order Books (2018, methodology): “…Regression (RR) [1], Single-Layer-Feedforward Network (SLFN) [1], Linear Discriminant Analysis (LDA) [22], Multilinear Discriminant Analysis (MDA) [22], Mul-tilinear Time-series Regression (MTR) [22], Weighted Mul-tilinear Time-series Regression (WMTR) [22], Multilinear Class-specific Discriminant…”
• Applications of deep learning in stock market prediction: recent progress (2020, unknown): “D CNN DNN, SVM Deng et al. (2019) KDTCN ARIMA, LSTM, CNN, TCN Sezer & Ozbayoglu (2019) CNN Buy&Hold Sim et al. (2019) CNN ANN, SVM Recurrent Neural Network Type Huynh et al. (2017) BGRU LSTM, GRU, DNN (Ding et al., 2014), DNN (Peng & Jiang, 2016) Li & Tam (2017) WT+LSTM LSTM Li et al. (2017a) LSTM S”