Stock Price Prediction Based on Markov Chains

Abstract

Short-term trend prediction in the stock market is of significant importance for effective market regulation by the government and optimizing resource allocation for investors. It has become a research hotspot in both academia and the industry in recent years. In addressing the long-term stock price prediction problem, a Markov Chain-based stock price prediction method is proposed. This method is based on the concept of state transitions in Markov Chains, where stock indicator data is transformed into state data. A transition probability matrix is generated, and predictions are made using matrix multiplication. Testing and verification are conducted using two datasets, BJ#430510 (Feng Guang Precision) and SZ#000009 (China Baoan). The results indicate that the stock price prediction model proposed in this paper exhibits high accuracy and stability.