Time series forecasting of stock market indices based on DLWR-LSTM model

Dingjun Yao and Kai Yan

Finance Research Letters, 2024, vol. 68, issue C

Abstract: Stock index forecasting is a hot research topic in the financial field. The traditional forecasting methods mostly use ARMA, ARIMA and GARCH to forecast the stock index. In recent years, many scholars have introduced machine learning such as SVM and RNN into the stock index forecasting model, but the accuracy of these forecasting results still needs to be improved. In this paper, by constructing DLWR-LSTM model, the trend of three indexes in Shanghai Stock Exchange is separated and predicted by layers to improve the accuracy of stock market index prediction, and the final prediction MAPE (average absolute percentage error) is close to 1 %. In this paper, different samples with different volatility but similar overall trend are replaced for experiments. The results show that the prediction accuracy of DLWR-LSTM model is not affected by the fluctuation of sample time series, and its final prediction result is maintained at around 1 % regardless of the variance of time series.

Keywords: Stock index forecast; Multiple trend separation; Hierarchical prediction; DLWR-LSTM model (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S1544612324008511
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:finlet:v:68:y:2024:i:c:s1544612324008511

DOI: 10.1016/j.frl.2024.105821

Access Statistics for this article

Finance Research Letters is currently edited by R. Gençay

More articles in Finance Research Letters from Elsevier
Bibliographic data for series maintained by Catherine Liu ().