论文部分内容阅读
金融市场是一个复杂、演化、非线性的动态变化的系统.金融数据往往带有噪声,非平稳且时常是混沌的.本文基于时序数据的先验知识——近期数据对于预测未来走势提供了更多的信息,对于传统的支持向量机的回归模型做出了一定的改进,即对于近期的数据预测错误施以更严重的惩罚,构建了改进的支持向量回归机模型.使用该改进模型对中国股票市场指数时间序列进行了预测,结果显示,本文改进的模型较之传统的支持向量回归机模型和神经网络模型有较好的预测效果.
Financial market is a complex, evolutionary and nonlinear system of dynamic change .Financial data is often noisy, non-stationary and often chaotic.This article is based on the prior knowledge of time-series data - recent data provide more information for forecasting future trends More information for the traditional support vector machine regression model to make some improvements, that is, for the recent data prediction error impose more severe penalties, build an improved support vector regression machine model.Using the improved model of China The stock market index time series are predicted. The results show that the improved model in this paper has a better forecasting effect than the traditional support vector regression model and neural network model.