基于时间序列方法的股票价格分析

Stock Prices Based on Time Series Analysis

摘要: 股票价格指数是一个国家经济建设健康状况的体温表，它的变化大致反映了该国经济结构和经济活动的宏观变化趋势。因此，选取合适的实证研究方法对股票价格指数进行预测分析，具有重要的实证意义。近年来，基于时间序列分析的预测方法在各个领域中都得到了广泛的应用。而对股票价格进行预测较为普遍的模型就是时间序列模型。所以本文以神州长城指数为例，将时间序列建模方法应用于股票价格指数的建模与预测，我们选择建立了ARIMA、GARCH模型进行拟合与预测。实验结果表明，该模型的残差白噪声检验，系数显著性检验都控制在一定范围内，因此该模型拟合效果较好，预测值接近实际值，最后，我们借助了该模型进行了股票指数未来一定时间内的预测。

Abstract: The stock price index is a thermometer of the health of a country’s economic construction. Its changes roughly reflect the macro trend of changes in the country’s economic structure and eco-nomic activities. Therefore, it is of great empirical significance to select appropriate empirical research methods to conduct predictive analysis of stock price indexes. In recent years, forecast-ing methods based on time series analysis have been widely used in various fields. The most common model for predicting stock prices is the time series model. Therefore, this article takes the Sino-Great Wall Index as an example to apply the time series modeling method to the mod-eling and prediction of stock price index. We choose to establish ARIMA and GARCH models for fitting and prediction. Experimental results show that the residual white noise test and coeffi-cient significance test of this model are controlled within a certain range, so the model has a good fitting effect and the predicted value is close to the actual value. Finally, we used this model to conduct stock index Forecasts within a certain period of time in the future.