股票市場(chǎng)的動(dòng)力學(xué)特征研究及智能預(yù)測(cè)模型.pdf

1、安徽大學(xué)碩士學(xué)位論文股票市場(chǎng)的動(dòng)力學(xué)特征研究及智能預(yù)測(cè)模型姓名：程蓉申請(qǐng)學(xué)位級(jí)別：碩士專業(yè)：運(yùn)籌學(xué)與控制論指導(dǎo)教師：毛軍軍201004安徽大學(xué)碩士學(xué)位論文股票市場(chǎng)的動(dòng)力學(xué)特征研究及智能預(yù)測(cè)模型IIAbstractNowadaysnonlineartheiessuchasfractalareappliedinFinancialCapitalSystem.Asthestockmarketisacomplexnonlinearsystemit

2、ismeimptanttodescribethetendencyexactly.Atthesametimepeoplepaymeattentiontothefecastofstockprice.Fexamplegreyfecastisappliedtofecaststockprice.Howeveritismuchimptanttoproperfecastmodeltoobtainthetrendofstockpriceinsuchtu

3、rbulentperiod.InthispaperweintegratewavelettheyfractaltheyfecastingtechnologytogetherapplytheminStockMarket.Themainwkresultsarefollowing:1.TheiesofRSMultifractalaregiventhetwomethodsareappliedinsixITcompanytoanalyzethete

4、ndencyofthestockprice.Meoverweanalyzethemultifractalfeatureofthestockpricenotonlyindinaryperiodbutalsointhefinancialcrisis.2.Basedonlinearregressionsupptvectregressionweconstructtwomodelswhicharewaveletlinearregressionmo

5、delwaveletSVMmodel.Thosetwomodelsbothcanvanishthenoisedisturbanceimprovethefecastingaccuracy.3.ThecombinationbasedonIOWAoperatiselabated.WeconstructanewmodelwhichcombineswaveletlinearregressionmodelwaveletSVMmodeltogethe

6、r.ComparedtoeitherwaveletlinearregressionmodelwaveletSVMmodelthenewmodelcansignificantlyimprovethefecastingaccuracy.Theeffectiveofthenewmodelisverified.Keywds:waveletanalysisRSanalysisMultifractalanalysislinearregression