【摘要】 针对马氏链短期预测股票价格的准确率问题,对状态缺失导致的马氏链预测模型失效问题进行了处理,基于滚动窗口下不同状态个数以及不同样本长度两个视角对股票价格数据进行动态处理,并分别对某股票每日收盘价未来1～5期所处的状态区间进行多次重复预测,最终将预测结果与实际值进行对比。研究结果表明,利用马氏链方法预测股票价格短期走势具有较高的准确率,准确率的高低随着状态个数、样本长度的不同而有所变动。更多还原

【Abstract】 The failure problem of Markov Chain prediction model caused by state loss is dealt with,based on the view of"number of different states"and "different sample length"under the rolling window,the data of stock price are processed dynamically,and then,respectively for a certain stock daily closing price in the future 1～5 interval repeat prediction of the state.Finally,the prediction results are compared with the actual values,and the accuracy of Markov Chain short-term prediction is analyzed.The results show that the Markov Chain method can predict the short-term trend of stock price with high accuracy rate,and the accuracy rate varies with the number of states and the length of samples.更多还原