应用财经新闻挖掘的金融品种价格走势预测

发布时间：2018-06-23 19:33

本文选题：财经新闻 + 金融市场预测　；参考：《计算机工程与科学》2016年09期

【摘要】：随着经济的不断发展,金融活动中的不确定性日益增加,金融预测受到学术界及金融界的高度重视。人们希望通过获得预测性的判断和推测,掌握金融产品未来的发展趋势和规律。而近期随着互联网发展,出现海量财经信息,仅仅依靠历史价格的数据挖掘技术,不能很好地反映金融市场多元因素的影响。因此,通过挖掘财经新闻信息中的情感倾向信息,结合金融历史价格数据,组合多元线性回归和差分自回归滑动平均模型,提出了一种基于财经新闻信息挖掘的金融价格走势预测方法,通过实际数据验证,表明该方法可以获得较为准确的预测结果。
[Abstract]:With the development of economy, the uncertainty in financial activities is increasing day by day. People hope to grasp the future trends and laws of financial products through predictive judgment and speculation. With the recent development of the Internet, the emergence of massive financial and financial information, relying solely on the historical price of data mining technology, can not well reflect the impact of multiple factors in the financial market. Therefore, by mining the emotional tendency information in the financial news information, combining with the financial historical price data, combining the multiple linear regression and differential autoregressive moving average model, A forecasting method of financial price trend based on financial news information mining is proposed. The actual data show that the method can get more accurate prediction results.
【作者单位】：国防科学技术大学计算机学院;
【基金】：国家自然科学基金(61379145,61170288,61272510)
【分类号】：F831.5;TP391.1

【相似文献】