Studying Stock Price Based On The Grey Markov Model

The stock market is an important part of the security market, in the final analysis people attach more attention to the stock price changes and the trend of the stock market in order to obtain short-term earnings with the deeply understanding on the stock market. Stock prices have a lot of volatility and uncertainty, and it is extremely difficult to master all information which affects the stock price, now there exists varieties of stock price forecasting models achieving good results, but the prediction model only based on the stock price is not much, of which the typical are the Grey model and the Markov chain model. Grey model mainly predict the stock market as a Grey system in which the stock price information is known and any other information is unknown. That is to say that the stock price in the stock market is a Grey value, the model predict the future shot-term stock price moves even the stock price specifically, according to the known stock price, which has the characteristics such as:little data modeling, simple in modeling, high accuracy and so on. However, the GM (1,1) model as the basis of the Grey Model prediction, which predicts the stock future price in the form of the specific values, and the solution of the model is the exponential curve, so it is not suitable for researching and forecasting the changeful stock price. Markov course is one species of stochastic course, and this Markov model expresses the Markov course in some specific form. It divides the known stock price series into differently finite states according to different standards, and then forecasts the future stock price states according to the transition probabilities between different states, reflecting the inherent law of the states. It is suitable for forecasting the changeful stock price series and sufficiently accurate stock price series. Thus, combining the two methods to make Grey Markov model prediction, which avoids the disadvantages of the single two method and carries forward the advantages of the both, improving the prediction accuracy obviously. The paper selects daily closing price index which reflects the overall trend of the Shanghai and Shenzhen A shares index from January4,2012to June13,2012as sample sequence, and establishes Grey Markov model for the stock price index, and predict the most possible value of the stock prices in the future five days. The result indicates that there exists error between the model prediction values and the actual values, but the relative error under this model is smaller and more accurate than the GM (1,1) model, which also shows that the model is more effective and superior for stock price forecasting and use the model to predict the stock price will provide partial reference and suggestions for the majority of investors.