期刊论文

A Hybrid Framework for Stock Price Prediction Based on PCA, ICEEMDAN, and LSTM

中文

应用数学进展

2324-7991

2023

12

5175-5185

本校为第一机构

投资者在市场上买卖股票的目的是为了获得最大的回报。然而,股票价格表现出非线性和非平稳性,难以准确预测。为了解决这个问题,结合主成分分析(PCA),完全自适应噪声集合经验模态分解(ICEEMDAN)和长短期记忆网络(LST...展开更多 投资者在市场上买卖股票的目的是为了获得最大的回报。然而,股票价格表现出非线性和非平稳性,难以准确预测。为了解决这个问题,结合主成分分析(PCA),完全自适应噪声集合经验模态分解(ICEEMDAN)和长短期记忆网络(LSTM),制定了一个混合预测模型,称为PCA-ICEEMDAN-LSTM,以预测中国股票指数收盘价。在这项研究中,选取8个股市中常用的技术指标作为原始特征,利用PCA筛选出最符...

The purpose of investors buying and selling stocks in the market is to achieve the maximum return. However, stock prices exhibit non-linearity and non-stationarity, making accurate prediction chal-lenging. To address this issue, a hybrid forecasting model, named PCA-ICE...MORE The purpose of investors buying and selling stocks in the market is to achieve the maximum return. However, stock prices exhibit non-linearity and non-stationarity, making accurate prediction chal-lenging. To address this issue, a hybrid forecasting model, named PCA-ICEEMDAN-LSTM, was de-veloped by combining Principal Co...