Applying Neural Networks to Prices Prediction of Crude Oil Futures英文文献资料.docVIP

HindawiPublishingCorporation MathematicalProblemsinEngineering Volume2012,ArticleID959040,12pages doi:10.1155/2012/959040 ResearchArticle ApplyingNeuralNetworkstoPricesPredictionof CrudeOilFutures andRickyRay-WenLin JohnWei-ShanHu,1,2Yi-ChungHu,2 3 1Department ofFinance,ChungYuanChristianUniversity,Chung-Li32023,Taiwan 2Department ofBusinessAdministration, ChungYuanChristianUniversity,Chung-Li32023,Taiwan 3TsannKuenTrans-NationGroup,Taiwan Correspondence shouldbeaddressedtoYi-ChungHu,ychu@.tw Received2March2012;Accepted 12June2012 Academic Editor:Jung-FaTsai Copyright q2012JohnWei-ShanHuetal. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction inanymedium,providedtheoriginalworkisproperlycited. The global economy experienced turbulent uneasiness for the past ?ve years owing to large increasesinoilpricesandterrorist’sattacks.Whileaccuratepredictionofoilpriceisimportantbut extremelydi?cult,thisstudyattemptstoaccuratelyforecastpricesofcrudeoilfuturesbyadopting threepopularneuralnetworks methods including themultilayer perceptron, theElmanrecurrent neuralnetworkERNN,andrecurrentfuzzyneuralnetworkRFNN.Experimental resultsindi- cate that the use of neural networks to forecast the crude oil futures prices is appropriate and consistent learning isachieved byemploying di?erenttraining times.Ourresultsfurtherdemon- strate that, in most situations, learning performance can be improved by increasing the training time. Moreover,theRFNN hasthebestpredictive power andtheMLP hastheworst oneamong thethreeunderlyingneuralnetworks.This?ndingshowsthat,underERNNsandRFNNs,thepre- dictive power improves when increasing thetraining time. The exceptional case involved BPNs, suggesting that thepredictive power improves when reducing thetraining time. Tosum up, we conclude that the RFNN outperformed the other two neural networks in forecas