ESET-ConvNet: Deep learning hybrid model for stock price prediction

Limit order book (LOB) tracks the outstanding limit order on a stock or any security. The LOB data is often used for high-frequency trading as it consists of detailed limit order information which cannot be presented from price data. Deep Learning (DL) is widely employed to train high-frequency trad...

Full description

Saved in:

Bibliographic Details
Main Author:	Ang, John Syin
Format:	Thesis
Published:	2022
Subjects:	Q300-390 Cybernetics
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my-mmu-ep.12038
record_format	uketd_dc
spelling	my-mmu-ep.120382024-01-11T01:28:19Z ESET-ConvNet: Deep learning hybrid model for stock price prediction 2022-09 Ang, John Syin Q300-390 Cybernetics Limit order book (LOB) tracks the outstanding limit order on a stock or any security. The LOB data is often used for high-frequency trading as it consists of detailed limit order information which cannot be presented from price data. Deep Learning (DL) is widely employed to train high-frequency trading models with LOB data because of its ability to extract features from unstructured data. There is lack of analysis between the existing models of time series and stock price prediction problems. This study has provided comprehensive analysis for time series modelling and stock price prediction problems. Besides, the existing stock prediction studies do not consider latent space of each stock. The aim is to find if learning the latent space for each stock can improve the model performance. This study has proposed a model architecture to combine and extend Temporal Convolutional Network (TCN), Squeeze-and-Excitation Network (SENet), and an embedding layer which is used for learning the latent space. Furthermore, the existing works have trained 5 stock sequences without splitting them to individual sequence as an input. In larger receptive field, this causes serious data leaking issue and this study have performed stock-splitting process to avoid the issue. The proposed model, ESET-ConvNet, has shown the improvement of F1- score. For 3 different prediction horizons, this model achieve the F1 score of 84.60%, 80.04%, and 83.77% respectively. The improvement of F1 score for each prediction horizons are 1.44%, 4.03%, and 4.26% respectively as compared to DeepLOB, the best performer among the LOB modelling studies. In the error analysis, it is realised that the distribution of classes for each stock and for each prediction horizon can be very different. In the ablation study, it is found that the embedding layer does not significantly improve the model performance. To further investigate the embedding layer, the suggestion of future work is to analyse the LOB models using a dataset with better coverage in terms of time span and number of stocks. 2022-09 Thesis http://shdl.mmu.edu.my/12038/ http://erep.mmu.edu.my/ masters Multimedia University Faculty of Computing and Informatics (FCI) EREP ID: 11733
institution	Multimedia University
collection	MMU Institutional Repository
topic	Q300-390 Cybernetics
spellingShingle	Q300-390 Cybernetics Ang, John Syin ESET-ConvNet: Deep learning hybrid model for stock price prediction
description	Limit order book (LOB) tracks the outstanding limit order on a stock or any security. The LOB data is often used for high-frequency trading as it consists of detailed limit order information which cannot be presented from price data. Deep Learning (DL) is widely employed to train high-frequency trading models with LOB data because of its ability to extract features from unstructured data. There is lack of analysis between the existing models of time series and stock price prediction problems. This study has provided comprehensive analysis for time series modelling and stock price prediction problems. Besides, the existing stock prediction studies do not consider latent space of each stock. The aim is to find if learning the latent space for each stock can improve the model performance. This study has proposed a model architecture to combine and extend Temporal Convolutional Network (TCN), Squeeze-and-Excitation Network (SENet), and an embedding layer which is used for learning the latent space. Furthermore, the existing works have trained 5 stock sequences without splitting them to individual sequence as an input. In larger receptive field, this causes serious data leaking issue and this study have performed stock-splitting process to avoid the issue. The proposed model, ESET-ConvNet, has shown the improvement of F1- score. For 3 different prediction horizons, this model achieve the F1 score of 84.60%, 80.04%, and 83.77% respectively. The improvement of F1 score for each prediction horizons are 1.44%, 4.03%, and 4.26% respectively as compared to DeepLOB, the best performer among the LOB modelling studies. In the error analysis, it is realised that the distribution of classes for each stock and for each prediction horizon can be very different. In the ablation study, it is found that the embedding layer does not significantly improve the model performance. To further investigate the embedding layer, the suggestion of future work is to analyse the LOB models using a dataset with better coverage in terms of time span and number of stocks.
format	Thesis
qualification_level	Master's degree
author	Ang, John Syin
author_facet	Ang, John Syin
author_sort	Ang, John Syin
title	ESET-ConvNet: Deep learning hybrid model for stock price prediction
title_short	ESET-ConvNet: Deep learning hybrid model for stock price prediction
title_full	ESET-ConvNet: Deep learning hybrid model for stock price prediction
title_fullStr	ESET-ConvNet: Deep learning hybrid model for stock price prediction
title_full_unstemmed	ESET-ConvNet: Deep learning hybrid model for stock price prediction
title_sort	eset-convnet: deep learning hybrid model for stock price prediction
granting_institution	Multimedia University
granting_department	Faculty of Computing and Informatics (FCI)
publishDate	2022
_version_	1794019132573745152

ESET-ConvNet: Deep learning hybrid model for stock price prediction

Similar Items