Research on Financial Time Series Prediction Based on Multi - Model Comparison

DOI

10.2991/978-94-6463-823-3_61How to use a DOI?

Keywords

Financial Time Series Prediction; Multi - Model Comparison; Evaluation Indicators

Abstract

In the digital age, artificial intelligence and finance are closely intertwined, and sophisticated models are constantly being developed and used extensively in the financial time series forecasting domain. On the other hand, prior research in this area had marked shortcomings. It is challenging to reliably ensure the model fitting degree, prediction accuracy, and research evaluation effect because the majority of prior research on financial time series forecasting just used one model or initial parameter without dynamic adjustment, and some studies only used one evaluation indicator. The purpose of this study is to forecast financial time series using a variety of methods, including comparison of several models, dynamic parameter adjustment and optimization, and thorough evaluation using a number of indicators. In order to create a comprehensive research system, this study chooses seven models, including Neural Networks (NNs), employs two dynamic parameter adjustment techniques, grid search and cross-validation, and uses five evaluation indicators, including the RMSE, based on the factors of comprehensiveness and efficacy. The work demonstrates how dynamic parameter modification, multi-model comparison, and thorough evaluation with different indicators may all greatly lower experimental errors while enhancing model fitting degree and prediction accuracy.

Copyright

© 2025 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  - Zixuan Ye
PY  - 2025
DA  - 2025/08/31
TI  - Research on Financial Time Series Prediction Based on Multi - Model Comparison
BT  - Proceedings of the 2025 3rd International Conference on Image, Algorithms, and Artificial Intelligence (ICIAAI 2025)
PB  - Atlantis Press
SP  - 615
EP  - 625
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-823-3_61
DO  - 10.2991/978-94-6463-823-3_61
ID  - Ye2025
ER  -