基于多状态马尔可夫链的股票价格预测
Prediction of Stock Price Based on Multi-State Markov Chain

随着人们投资理财的意识的逐渐建立，股票已成为常见的投资对象之一。但是在股票带来高收益的同时伴随着高风险。如果能提供较准确的股票参考预测，对投资者以及企业自身都将得到可观的收益。由于影响股票市场的因素众多，因此股票波动具有一定的随机性。本文围绕股票只受当前状态的影响、与过去状态无关的特点，将运用广泛的马尔可夫链作为预测模型。以在线旅游板块中的西域旅游股票作为案例，并对其进行运算处理，建立四状态和多阶十六状态马尔可夫链模型，预测其发展趋势，得出马尔可夫链在股票的实际应用中具有一定研究价值的结论。
With the gradual establishment of people’s awareness of investment and financial management, stocks have become one of the common investment objects. But with high returns come high risks. If we can provide more accurate stock reference forecasts, we will get considerable benefits for investors and companies themselves. Because of the many factors that affect the stock market, stock fluctuations are somewhat stochastic in nature. In this paper, around the feature that stocks are only affected by the current state and have nothing to do with the past state, the widely used Markov chain is used as a prediction model. The Western Region Tourism stock in the online tourism sector is used as a case study and processed arithmetically to establish a four-state and multi-order sixteen-state Markov chain model to predict its development trend, and conclude that Markov chain has some research value in the practical application of stocks.