證券價(jià)格的預(yù)測(cè)方法研究.pdf

1、南京信息工程大學(xué)碩士學(xué)位論文證券價(jià)格的預(yù)測(cè)方法研究姓名：胡俊勝申請(qǐng)學(xué)位級(jí)別：碩士專業(yè)：系統(tǒng)分析與集成指導(dǎo)教師：肖冬榮20050501AbstractThedevelopmentofourcountrysstockmarketisfast.Theforecastingmethodsofsecuritymarketarevarious.Themethodofquantitativeanalysisisconsistsofthelineara

2、ndnonlinearmethod.Thelinearmethodbeusedtoforecastthestockmarketpricewhichoftenhassomelocalization.Nowthenonlinearmethodismoreandmoreabundantwhichhasbeenshowitsadvantageinstockmarket.Thisthesissimplyintroducesthelinearand

3、nonlinearmethodandtheirpresentconditionsexpoundingtheresearchstatusoftheneuralnetworkandthewaveletneuralnetwork.ThenwesetupthelineartimeseriesmodeltoforecasttheclosingpriceoftheChangJiangelectricpower.Theneuralnetworkhas

4、mightynonlinearability.ItwasbeenwidelyusedinmanyFields.ThethesisusestheBackPropagationmethodwhichisfrequentoftheneuralnetworkbuildingthethreelayernetworkfortheforecastingoftheclosingpriceofChangJiangelectricpower.Finally

5、accordingtoclosingpricecurvethewaveletneuralnetworkwassetuptoforecastit.Intheprocessofdesigningthewaveletneuralnetworkthethesisdiscussesthewaytoinitializetheparameterofthenetworkandusestheselfrealizationstepmethodtothest

6、udyprocessofwaveletneuralnetwork.Withthecomparisonofthesethreekindsofforecastingmethodsthethesismakesasumup.Intheseforecastingmethodsthenonlinearmethodoftencangetabeterforecastingaccuracythanthelinearmethod.Thewaveletneu

7、ralnetworkhasacharacteristicofselftraining.Ithasfasterconvergentspeedandhigherprecisionthanthetraditionalneuralnetwork.Ithasgoodforegroundinthenonlinearforecastingfields.KeyWords:ForecastingArimamodelNeuralNetworkWavelet