CRUDE OIL PRICE FORECASTING WITH TEI@I METHODOLOGY

资源类型:pdf 资源大小:1.93MB 文档分类:经济 上传者:王敏
关闭
关闭
关闭
关闭
关闭
关闭
关闭
关闭
关闭
关闭

最多预览10页,更多请下载原文查看

文档信息

【出版日期】2005-04-10

【摘要】<正>The difficulty in crude oil price forecasting, due to inherent complexity, has attracted much attention of academic researchers and business practitioners. Various methods have been tried to solve the problem of forecasting crude oil prices. However, all of the existing models of prediction can not meet practical needs. Very recently, Wang and Yu proposed a new methodology for handling complex systems-TEI@I methodology by means of a systematic integration of text mining, econometrics and intelligent techniques. Within the framework of TEI@I methodology, econometrical models are used to model the linear components of crude oil price time series (i.e., main trends) while nonlinear components of crude oil price time series (i.e., error terms) are modelled by using artificial neural network (ANN) models. In addition, the impact of irregular and infrequent future events on crude oil price is explored using web-based text mining (WTM) and rule-based expert systems (RES) techniques. Thus, a fully novel nonlinea

【刊名】Journal of Systems Science and Complexity

全文阅读

1 Introduetion 011 pl盯5 an inereasingly signifieant role in the world eeononly sinee nearly two-thirds of the world’5 ener罗eonsumption eomes from erude 011 and natural gas【i}.worzdwide eonsumption of erude 011 exeeeds$500 billio一roughly 10%of US GDP,and erude 011 15 also the world,s lar罗st and most aetively traded eonunodit弘aeeounting for about 10%of total wor一d trade[2]. 146 认城NG SHOU、叭NG et al. 从)1 .18 Crude 011 15 traded internationally among many different PI盯ers一011 Produeing nations,011 eomPanies,individual refineries,011 imPorting nations and 011 sPeeulators.The erude 011 Priee 15 basieally determined场its supply and demand,and 15 strongly influeneed场mally irregular past/present/future events like weather,stoek levels,GDP growth,politieal asPeets and people,s exPeetations.Furthermore,sinee it takes eonsiderable time to shiP erude 011 from one eountry to another,011 priees vary in di价rent parts of the world.These faets lead to a strongly fluetuating and interaeting market and the fundalnefltal meehanism governing the eomPlex dynamie 15 not understood.As Mauriee[3]reported,the 011 market 15 the most volatile of all the mar挽ts exeept Nasd叫.In addition,as sharP 011 Priee movements are likely to disturb aggregate eeonomie aetivity,volatile 011 priees are of eonsiderable interest to ma即researchers and institutions. Therefore,foreeasting 011 Priees 15 a very imPortallt toPie,although it 15 an extremely hard one due to its intrinsie diffieulty and Praetieal apPlieations.Crude 011 Priee eh汕ges may Potentially have signifieant effeets on the eeonomie Performanee of both 011 imPorting and 011 exPorting nations.The main reasons are tw于folds.On one hand,sharP inereases of erude 011 priees adversely affeet the eeonomie growth and inflation in 011 imPorting eeonomies.On the other hand,erude 011 Priee falls,like the one in 1998,ereate serious budgetary Problems for 011 exporting eountries[4]. When it eomes to erude 011 Priee foreeasting,most of the literatures foeus only on 011 Priee volatility analysis(see e.g.!l〕,!5})and 011 priee determination within the supply and demand framework(see e.g.【6,7」).There 15 only very limited researeh on 011 priee foreeasting,in- eluding quantitative and qualitative methods,Among the quantitative methods,Huntingtonls] used a soPhistieated eeonometrie model to foreeast erude 011 Priees in the 19805.Abrarnson and Finizza【9]utilized a probabi一istie model for芦edieting 01一priees,and Barone一Adesi et al.110! su既ested a semi一parametrie approaeh for foreeasting 012 priee.Moranalll]used the samea卜 Proaeh to foreeast short一term 011 Priees and rePorted some Performanee imProvement.Among the qualitative methods,Nelson et al.[12}used the oelphi method to foreeast 011 priees for the california Energy commission.A玩amson and Finizza〔131 used belief networks,a elass of knowledge一based models,to foreeast 011 Priees.However,the above methods ean,t meet Praet卜 eal needs in foreeajsting erude 011 Priees.Henee,it 15 of signifieanee to develoP new foreeasting methods of foreeasting 011 Priees. In view of the great diffieulty and eomPlexity of erude 011 Priee foreeasting,in this studywe apply a new methodology named TEI@1.From the name TEI@I,one ean easily understand that the methodology 15 based on the idea of text mining+eeonometries+intelligenee(intelligent algorithms)@integration,Using“@”to replaee“+,,15 to emphasize the eentral role and fune- tion of integrations.In this methodology,eeonometrieal models(e.g.,autoregressive integrated moving average(ARIMA))are used to model the linear eomponents of erude 011 priee time series(i.e.,main trends)while nonlinear eomponents of erude 011 priee time series(i.e.,error terms)are modeled by using an artifieial neural network(ANN)model.In addition,thee瓜ets of irregular and infrequent events(p韶t/present/future)on erude 011 priee are explored by using web一b、ed text mining(WTM)and rule一based expert systems(RES)teehni卿es.Thus,within the framework of TEI@1 methodolog弘a novel nonlinear integrated foreeasting aPProaeh with error eorreetion and judgmental adjustment 15 formulated to have aeeurate Predietions and im- Proving foreeasting performanee.Toe却lain the methodology and to veri勿the effeetiveness of the novel foreeasting aPProach,a simulation stu勿with Practieal data 15 Presented. The rest of the paper 15 organized as follows.Seetion 2 presents a general deseription of TEI@1 methodology.A nonlinear integrated foreeasting aPproaeh with error eorreetion and judgmental adjustment within the framework of TEI@1 methodology 15 ProPosed in this seetion.To evaluate the ProPosed apProach,an emPirieal study 15 Presented in Seetion 3.Some eoneluding remarks are made in Seetion 4. No.2 CRUDE OIL PRICE FORECASTING WITH TEI@1 METHODOLOGY 147 2 TEI@IMethodologyforCrude0ilPrieeForeeasting 玩this seetion,a general deseriPtion of the TEI@1 methodology 15 Presented via the appli- eation of erude 011 foreeasting.Firstl又we deseribe the frarnework of TEI@1 methodology,and then diseuss some m

1 2 3 4 5 6 7 8 9 10 11 12 13

引证文献

问答

我要提问
关闭