论文部分内容阅读
利用小波分析预测方法对金融数据—股票收盘价这一典型的非平稳时间序列进行预测.使用M a llat小波分解算法对数据进行分解,对分解后的数据进行平滑处理,然后再进行重构,而重构之后的数据就成为近似意义的平稳时间序列,这样就得到了原始数据的近似信号,再应用传统时间序列预测方法对重构后的数据进行预测,将预测结果与实际值,以及和传统预测方法预测结果比较,小波分析方法预测效果更为理想.
This paper forecasts the typical nonstationary time series of financial data-stock closing price by using wavelet analysis and forecasting method, decomposes the data by using Malatlat wavelet decomposition algorithm, smoothes the decomposed data and reconstructs it, However, the reconstructed data becomes a stationary time series of approximate meaning, so that the approximate signal of the original data is obtained. Then the traditional time series prediction method is used to predict the reconstructed data, and the prediction result and the actual value, and Compared with the prediction results of traditional prediction methods, the prediction effect of wavelet analysis method is more ideal.