基于时间序列分析方法的股票价格预测——以平安银行股票为例  

Stock Price Prediction Based on Time Series Analysis Method—Taking Ping An Bank Stock as an Example

作　　者：付春蕊 

机构地区：[1]南京信息工程大学数学与统计学院,江苏 南京

出　　处：《运筹与模糊学》2023年第6期7283-7293,共11页Operations Research and Fuzziology

摘　　要：随着金融行业的蓬勃发展,股市的热度随之增高。面对日益繁荣的股市发展,现如今对于投资决策的研究日益重要,如何对股票价格的准确预测更是核心关键。影响股票价格波动的因素类型繁多,包括国家制定的经济政策、公司运营状况、市场经济发展状况等,因为这些随机性因素的存在,使得股价预测难以达到精准,只能在有限范围内做出最佳预测。时间序列分析是经济领域中用来拟合金融数据变化的重要工具。在股票市场中,时间序列分析常用于预测股票价格随时间的变化趋势,具有一定的适用性,能够在一定范围内对股票价格波动做出有价值的预测。本文主要介绍了时间序列分析方法的定义,性质以及建模过程。对平安银行股票实证研究分析:建立时间序列模型,具体包括ARIMA模型、残差自回归模型,对股票每日开盘价进行探究。最后通过拟合效果和预测效果比较,对比了不同时间序列分析模型的优劣,探讨了其对于股票预测实施的参考价值。As the financial industry booms, the stock market becomes more popular. Facing the increasingly prosperous development of the stock market, the research on investment decisions is becoming increasingly important, and how to accurately predict stock prices is the core key. There are many types of factors that affect stock price fluctuations, including economic policies formulated by the state, company operations, market economic development, etc. Because of the existence of these random factors, it is difficult to achieve accurate stock price predictions, and the best predictions can only be made within a limited range. Good prediction. Time series analysis is an important tool in the economic field used to fit changes in financial data. In the stock market, time series analysis is often used to predict the changing trend of stock prices over time. It has certain applicability and can make valuable predictions on stock price fluctuations within a certain range. This article mainly introduces the definition, properties and modeling process of time series analysis methods. Empirical research and analysis on Ping An Bank’s stocks: Establish a time series model, specifically including ARIMA model and residual autoregressive model, to explore the daily opening price of the stock. Finally, by comparing the fitting effect and prediction effect, the advantages and disadvantages of different time series analysis models were compared, and their reference value for the implementation of stock prediction was discussed.

关 键 词：股票价格 时间序列分析 预测 

分 类 号：F83[经济管理—金融学]