基于隐马尔可夫模型的股票价格预测

Stock Price Prediction Based on Hidden Markov Model

摘要: 本文构建隐马尔可夫模型预测比亚迪公司股票收盘价，采用K均值聚类法和AIC、BIC准则确定隐状态个数，运用EM算法进行模型参数估计，并将MSE、MAE和R²作为评价指标评估准确性，结果显示基于模型预测结果较为准确稳定。研究结果表明HMM模型能捕捉市场因素、公司财务状况和行业趋势对价格的影响，为投资者和分析师提供深入市场洞察。本研究提供了有效的股票预测模型，同时探索了HMM模型在股票价格预测中的应用，为金融时间序列预测方法的改进和发展提供新思路和方法。

Abstract: In this paper, a hidden Markov model is constructed to predict the closing price of BYD Company’s stock, the number of hidden states is determined by K-means clustering method, AIC and BIC criteria, and the model parameters are estimated by EM algorithm. MSE, MAE and R² are used as evaluation indicators to evaluate the accuracy. The results show that the prediction results based on the model are more accurate and stable. The results show that the HMM model can capture the impact of market factors, company financial conditions and industry trends on prices, providing investors and analysts with in-depth market insights. This study provides an effective stock prediction model, and explores the application of HMM model in stock price prediction, which provides new ideas and methods for the improvement and development of financial time series prediction methods.