Technology Stock Forecasting Based on Hybrid Model and Data Migration

DOI

10.2991/978-94-6463-304-7_41How to use a DOI?

Keywords

Hybrid model; ARIMA; Deep Learning; Data Migration

Abstract

The forecasting of financial time series has long been recognized as a challenging and important task for both economists and computer scientists. In recent years, the integration of deep learning and artificial intelligence has significantly advanced the field of financial time series forecasting. Deep learning has been demonstrated to yield superior results in handling nonlinear trends, which are difficult to model using traditional linear models. In this study, we propose a two-part method that integrates linear and deep learning models to forecast financial time series. In the first part, we utilize the Autoregressive Integrated Moving Average (ARIMA) model to filter the linear trend of the stock and pass the residual value to the second part. In the second part, we employ a Convolutional Neural Network (CNN)-Bidirectional Gated Recurrent Unit (BIGRU) neural network with an attention mechanism to effectively process the residual value and make predictions. The dataset we selected consists of stocks from the technology sector, and there is a certain similarity in the trends of these stocks. To improve performance, we use data migration to improve its performance. To evaluate the proposed model, we use Mean Squared Error (MSE) and Mean Absolute Error (MAE) to measure its performance. We compare our proposed method with a benchmark approach, and the experimental results demonstrate that our method has higher prediction accuracy. Our approach thus presents significant advantages for forecasting financial time series.

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  - Yong Lu
AU  - Lumiao Hu
PY  - 2023
DA  - 2023/12/04
TI  - Technology Stock Forecasting Based on Hybrid Model and Data Migration
BT  - Proceedings of the 3rd International Conference on Digital Economy and Computer Application (DECA 2023)
PB  - Atlantis Press
SP  - 383
EP  - 394
SN  - 2589-4900
UR  - https://doi.org/10.2991/978-94-6463-304-7_41
DO  - 10.2991/978-94-6463-304-7_41
ID  - Lu2023
ER  -