A Study of Stock Price Function Based on Hybrid Deep Learning Model

DOI

10.2991/978-94-6463-300-9_5How to use a DOI?

Keywords

Hybrid Model; Deep Learning; Transformer; Unsupervised Learning; Price Function

Abstract

Recently, global economy has recovered from the COVID impact. Stock index is reasonable indicator of economy. The prediction of stock index contributes to both forecasting trend of global economy and portfolio construction. According to the characters of price factors, this paper proposes a hybrid model, combining Variational Autoencoder (VAE), Multi-directional Delayed Embedding (MDT), Gated Recurrent Unit (GRU), and Multi-Head Self-Attention (MHSA) modules, named VMGM, to predict the movement of Shanghai Stock Exchange (SSE 50). It combined deep learning, unsupervised learning, and transformer to one model whose prediction result is more accurate than traditional model. The original data contain 7 price factors of SSE 50 from 2004 to 2020. Writer uses VAE and moving average convergence/divergence (MACD) to generate new price factors for preprocessing and feature engineering. Final price factors are transformed by MDT and convolutional neural network (CNN) before entering GRU and Multi-Head self-attention. The prediction result of model is more accurate than controlled sets which prove the choice logistic of each module.

Copyright

© 2023 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Yangwenyuan Deng
PY  - 2023
DA  - 2023/11/27
TI  - A Study of Stock Price Function Based on Hybrid Deep Learning Model
BT  - Proceedings of the 2023 International Conference on Image, Algorithms and Artificial Intelligence (ICIAAI 2023)
PB  - Atlantis Press
SP  - 35
EP  - 42
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-300-9_5
DO  - 10.2991/978-94-6463-300-9_5
ID  - Deng2023
ER  -