互联网股市信息强度的统计分类及其在股价波动上的预测.pdf

计算机研究与发展 ISSN1000—1239／CN11-1777／TP of Researchand JournalComputer Development 42(增刊)：158～162，2005 互联网股市信息强度的统计分类及其在股价波动上的预测 梁 循 (北京大学计算机科学与技术研究所北京 100871) of Scienceand Alto，CA94305，UsA) (DepartmentManagement Engineering，StanfordUniversity，Palo (1iangxun@icst．pku．edu．cn) ClassificationofWebStockNewsIntensitiesBasedonStatisticsandPredictionon ofStockPrices Changes Xun Liang (Institute Scienceand 100817) ofComputer Technology，PekingUniversity，Beijing Scienceand Alto，CA94305，USA) (DepartmentofManagementEngineering，StanfordUniversity，Palo AbstractThe increasesofwebstocknews oftenassociatedwiththe significant intensities(WSNI)are ofstock theWSNIfora increases significantchanges prices．When company insufficiently．the stock movementisoften theWSNIincreases correspondingprice placid．When significantly．the stock movementisoften measureof increasesofWSNIis correspondingprice oscillatory．Thesignificant measureof ofstock is showthatthe defined．Also，a significantchanges prices established．Experiments classificationsare tofindtheexistenceofrelationsbetweentheWSNIsandstock helpful prices。and enhancethe accuraciesofWSNIsonthestock aresult。thenewcluesareaddedtothe prediction prices．As microstructureofthestockmarketfromthe oftheInternetnews． angle wordsInternet；Webstocknews Key intensities；stockprice；classification；prediction 摘要互联网股市信息强度的显著增加常常和股价的显著变化相关联．当某公司的这个