Research Article
Stock Change Predictions with ARIMA Model
@INPROCEEDINGS{10.4108/eai.7-7-2023.2338052, author={Zihao Guo}, title={Stock Change Predictions with ARIMA Model}, proceedings={Proceedings of the 2nd International Conference on Financial Innovation, FinTech and Information Technology, FFIT 2023, July 7--9, 2023, Chongqing, China}, publisher={EAI}, proceedings_a={FFIT}, year={2023}, month={10}, keywords={stock prediction; the arima time series model; the multiple linear regression model}, doi={10.4108/eai.7-7-2023.2338052} }- Zihao Guo
Year: 2023
Stock Change Predictions with ARIMA Model
FFIT
EAI
DOI: 10.4108/eai.7-7-2023.2338052
Abstract
The Stock investment is a wealth management model, based on the customer's financial situation, financial goals and risk preferences and other factors for customers to generate intelligent and personalized investment solutions. However, algorithm for predicting stock changes is complex. And wide range of equity investment options based on different algorithms and factor indicators. Therefore, choosing options for investors is difficult. To address the wide disparity between different equity investment options, the present study proposes that forecasting analysis of stocks based on ARIMA time series models. First, the previous stock data will be filtered, and the ARIMA time series model will be used to forecast the changes The results of this study showed that compared to the multiple linear regression prediction model that is based on ARIMA time series model of the Stock prediction, the prediction root mean square error decreased by 25%, the correlation coefficient increased by 15%. Therefore, ARIMA time series model can help the investors to choose the Stock exactly, and it has important role in stock price forecasting.