Technical Analysis and market efficiency

《Technical Analysis and market efficiency》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。

TechnicalAnalysisandmarketefficiency:AstudyofUKmarketAbstractThepurposeofthedissertationistostudytherelationshipbetweentechnicalanalysisandtheefficientmarkethypothesis.Toempiricallytestwhethermarketinefficiencyisassociatedwiththeinformationofpositivefeedbacktechnicaltradingrules,thisdissertationfocusesontheUKstockmarket,inwhichtechnicalanalysisiswidelyused.Thedatacoversthedailyclosingvaluesoftop30stocksintheFTSE100indexfrom1987–2010.Fourapproacheswillbeusedinthisdissertationtostudytherelationshipbetweentechnicalanalysisandmarketinefficiency. ContentsChapter1IntroductionandStatementofProblem3Chapter2LiteratureReview4Chapter3Methodology8References9 Chapter1IntroductionandStatementofProblemTheefficientmarkethypothesis(EMH)hasbeenwelladoptedinfinancialliteraturebytheacademia.TheassumptionsofEMHare:(1)investorsbehaverationally,(2)intrinsicvalueofstockpricesarereflectedimmediatelyand(3)investorsreactpromptlytoreflecttheintrinsicvalue.Thisdissertationaimsatansweringtwoquestions.Firstofall,marketpractitionershavebeenfindingwaystopredictstockspricesinthefuturebyapplyinganalysisofhistoricaldata.Thisschoolofthoughtiscalledtechnicalanalysis.ProminentexamplesoftechnicalanalysistheoriesincludetheDowTheory,ElliottWaveTheoryandGannTheory.Allofthesetechnicalanalysistechniquesrestsoftheassumptionthatstockpricesdonotreflectintrinsicvalueandispredictable.Thebuildingblocksofmodernfinancetheory,whicharebasedonefficientmarkethypothesis,remainquestionablebyfinancepractitioners.Financepractitionersseekforforecastingmodelsandtradingstrategieswhichcangenerateprofitsfromthecapitalmarket.Thevolumeofliteraturediscussingefficientmarkethypothesishasoutweighedsimilarstudiesinacademia.Thedissertationaimsatprovidingempiricalevidencestoshowthatwhethertechnicalanalysiscanbeafactortoexplainmarketanomalies.Empiricalanalysiswillbeperformedtodecidewhetherthetechnicalanalystcanoutperformthemarketinstatisticalsense.Thisdissertationservesthepurposeofbridgingthegapbetweenthesetwogroupsbygivinganassessmentoftechnicalanalysis.Thewidespreaduseoftechnicalanalysisinprofessionalarenahastriggeredstudiesontheimpactofthesetechnicaltradingrulestomarketprices.Theoverwhelmingpopularityoftechnicalanalysisinthepublicandprofessionalarenaappearstoviolate theexpectationoffinancialtheoriststhatsuchmethodsshouldbeproveduselessbyinvestors.Chapter2LiteratureReviewThischapterisdividedintotwomainsections.Thefirstsectionofthischapteristoreviewexistingliteratureindiscussingefficientmarkethypothesisandapproachesadoptedbypreviousstudiesinassessingcriticallyinefficientmarkethypothesis.Thesecondsectionofthisdissertationistogiveabriefreviewonthescopeoftechnicalanalysisinbothacademicliteratureandotherliterature,forinstance,Pring(1991).Theefficientmarkethypothesishasbeenestablishedforseveraldecades.TheEMH,proposedbyFama(1965,1970,1991)andothers,assertsthatassetpricesshouldpromptlyreflectallrelevantmarketinformation.TherearethreeformsofmarketefficiencystatedinEMH.Fama(1970and1976)hasdoneagreatjobbydefiningthreeformsofEMH.Theyare:(1)Effectivemarketandtechnicalanalysis.(Weakformofmarketefficiency)Ifthemarketdoesnotmeetconditionsundertheweak-formofmarketefficiency,thecurrentpriceisnotfullyreflectthehistoricalpriceinformation,futurepricechangeswillfurtherpriceinformationtorespondtothepast.Inthiscase,peoplecantakeadvantageoftechnicalanalysisandchartingpastpriceinformationintheanalysisofthefuturepriceofsomesortofchange,resultinginaprofitinthetransaction.Ifthemarketisweakandeffective,thenthepasthistoricalpriceinformationisfullyreflectedinthecurrentprice,pricechangesforthefuturehasnothingtodowiththecurrentandhistoricalprice,usingtechnicalanalysisandchartinganalysisofcurrentandhistoricalpriceforecastsforthefuturewillbeinvain.Ifyoudonotusefurtherinformationotherthanthesequenceofprices,pricepredictionvalueswillbethebest tomorrowistoday'sprices.Weak-formefficientmarket,technicalanalysisisinvalidated.(2)Effectivemarketandfundamentalanalysis.(Semi-strongformofmarketefficiency)Ifthemarketnotyetreachedconditionsforsemi-strongformofmarketefficiency,publiclyavailableinformationisnotcurrentpricesfullyreflect,analysisofaccesstoinformationtofindfalsepositivesfixedpricewillbeabletoincreaseearnings.However,ifthemarkethalf-effective,thentheonlyanalysisbasedontheaccesstoinformationwillnotbeabletoprovideanyhelp,becausetheinformationyoucurrentlyhaveopen,thecurrentpriceissuitable,futurepricechangesofthecurrentlyknownpubliclyavailableinformationaboutnothingtorelypurelyontheirchangetomorrowanewpublicinformation.(3)EffectivemarketandPortfolioManagement.(Strongformofmarketefficiency)Ifthemarketisefficientinstrong-form,andaccesstointernalinformationandfollowtheaction,inwhichcaseanynewinformation(includingpublicandinternal)willbequicklyreflectedinthemarket.Sointhismarket,anyattempttolookforinformationtocrackdownonmarketpracticesarenotwise.Underthiskindofstronglyefficientmarketshypothesis,anyprofessionalinvestormarginalmarketvalueiszero,becausethereisnosourceofinformationandtheprocessingmodetoprovideasteadyincreaseinearnings.Portfoliotheory,portfolioconstructionisoneoftheconditionsistoassumethatthestockmarketisfullyeffective,allmarketparticipantshaveequalaccesstoadequateinvestmentinformation,suchasstockgainsandriskfactorsofchangeanditseffectswithouttransactioncosts.Forportfoliomanagement,but,ifthemarketisstrongandeffective,andportfoliomanagersselectthenegativeattitudeof theconservative,justtogetthelevelofaveragemarketrateofreturn,becausethedifferencebetweenafutureperiodofsupportiveandnon-profitsinvestmentscannotbeknownatthisstageofanyfeaturesoftheseinvestmentsonthebasisofcombinedadjustment.DeBondandThaler(1985,1987)1926~1982inNYSE-listedstocksduringtheyearstudyfoundthatthebest35stocksoverthepast5years(winnercombinations)andworst-performing35stocks(loserscombination)earnings,wasreversedinthenext3years,"loserscombination"averagecumulativeearningspershare"winnercombinations"at25%.ConradandandKaul(1988)andLoandMackinlay(1988)studiesfound,weeksofstockreturnsareseriallycorrelated,butthecorrelationcoefficientsarequitesmall,thereisnoclearindicationoftradingopportunitiesexist.Insubsequentstudies,JegadeeshandTitman(1993)demonstrated,foraperiodofonemonthorthefollowinginvestmentinthereverseeffect.Chopra,andLakonishokRitter(1992)isconfirmDeBondandThaler(1985,1987)found.JegadeeshandTitman(1993)studyfoundthattheearningsperformanceofthestockduringtheperiodinthe3~L2monthsforserialcorrelation,inertialeffect,ifin6months-longinvestigationofstockreturns,"winnersportfolio"and"whatloserscombination"averageearningsbyaround9%.Chan,JegadhesshandLakonishok.(1996),JegadeeshandTitman(1993)gotthesameresultsofthestudywithanenlargedsample.Rouwenhorst’s(1998)studyshowedthatinotherdevelopedmarkets,thereareinertiaeffectsaswellassomeemergingstockmarkets.MoskowitzandGrinblatt(1999)examinedtheinertialeffectsofportfoliobyindustryclassification,foundintheUnitedStatesstockmarket,significantinertiaeffectsincombination,andthecombinationofexceptionalearningspersharemore.PoterbaandSummers(1988),LoandMacKinlay(1988,1989)alsofoundevidenceonmeanreversions.Besides,somestudiesfindthepresenceofmarketoverreactionandtheprofitabilityofcontrarian tradingrules.AcomprehensivereviewontestingEMHcanbefoundinReiliy(1994)Mostoftechncialtradingrulesaredesignedasmomentumindicators.Momentumindicatorsaredesignedtofollowtrends.Thistrend-followingapproachhasbeeninvestigatedintheliteraturebyTverskyandKahneman(1982),Kroll,LevyandRapport(1988),DeBondt(1993)andFalkenstein(1996).Findingsintheseexperimentalstudiesdiscoveredthatinvestorshavetendencytoextrapolatetrends.DeLong,Shleifer,SummersandWaldman(1990)haveprovidedaframeworkonthisrationalalebasedonnoisetraderrisk.Noisetradersfollowpositivefeedbacktradingstratgiesandshort-termrationalspeculatorsonpopularinformation,itiscrucialtostudytheinfluenceoftechnicalanalysisonpricebehavior.Therearetworeasons.Firstofall,manytradersusetechnicalanlaysisandmosttechnicaltradingmethodsrelyontrend-chasingmethodology.Trendchasingisessentiallyapositivefeedbacktradingstrategy.Asaresult,iftechnicalanalysisispopularinamarket,positivefeedbackbehaviorshouldbepronouncedinsuchmarket.Technicalanalysishasbeendevelopedandadoptedbyprofessionaltradersformorethanacentury.Pring(1991)hasacomprehensivereviewontechnicalindicatorsusedbyprofessionaltradersinthemarket.Ontheotherhand,therehaveevidencesthattechnicalanalysishasbeenwelladoptedbyprofessionaltraders.Forexample,AllenandTaylor(1989)findthatmajorityofforeignexchangedealersinbanksusetechnicalanalysisandtheimportanceoftechnicalanalysiscarriessimilarweightingasfundamentalanalysisintheirdecision-makingprocess.CarterandVanAuken(1989)showthattechnicalanalysishasbeenwidelyusedbytradersintheU.S.market.FrankelandFroot(1986)reportthattheuseoftechnicalanalysisamongprofessionaltradersasforecastingtoolsarecommoninforextradingin1980s.Paststudiesontechnicalanalysisaremostlyconcernedwithfilterrules.Theyusually simulatetradingrulesandtesttheirperformanceintermsofriskandreturn.Moststudiesinthisdirectionsuggestthattechnicalanalysisisoflitterusebasedonthefactthatabnormalreturnscannotbegeneratedaftertransactioncostsbyapplyingabuy-and-holdstrategy,asdocumentedinFamaandBlume(1966),Fama(1970),Levich(1985and1989).Ontheotherhand,mostofthesestudiesfocusedontheU.S.stockmarket.Thesetechnicaltradingrulesmaygenerateoutperformancewithreferencetobenchmarkinothermarketsandotherassetclasses.Chapter3MethodologyThisdissertationattemptstotestempiricallyhowtechnicaltradingrulescanbeusedtolinkwithmarketinefficiencies.Inthisdissertation,tradingsignalsaregeneratedbasedonRelativeStrengthIndex(RSI).RSIisselectedastheproxiesfortheinformationoftechnicalanalysis.Thesefiguresarereadilyavailableinmediaandfinancialwebsitesandmarketdataprovidersofferreal-timedataonthesefiguresandthesearehighlyrecommendedbytechnicalanalysts.Iftheuseoftechnicalanalysisiswidelyadoptedbymarketparticipants,itislikelythatsharedbeliefcanstronglyinfluencepriceswhichcanbeusedasevidencesformarketinefficiency.TotesttherelationshipbetweenRSIrulesandmarketinefficiency,thefollowingapproachesareadopted1.Investigatingtheassetpricebehaviorshortlyafterthesesignalsaregenerated2.Testingtheshort-termautocorrelationofreturnsshortlyafterthesesignalsaregenerated3.Testingthemean-revertingpatternofassetpricesafterthesesignalsaregenerated4.Assessingtheportfolioperformancebyapplyingthesetechnicalsignals. ReferencesAllen,H.andTaylor,P.(1989)“ChartAnalysisandtheForeignExchangeMarket,”BankofEnglandBulletin,November,547-551.Blume,L.,Easley,D.,andO’Hara,M.(1994)“MarketStatisticsandTechnicalAnalysis:TheRoleofVolume,”JournalofFinance,153-181.Brock,William,JosefLakonishok,andBlakeLeBaron(1992)"SimpleTechnicalTradingRulesandtheStochasticPropertiesofStockReturns,"JournalofFinance,47:1731-1764.Brown,D.,andJennings,R.(1989)“OnTechnicalAnalysis:TheRoleofVolume,”ReviewofFinancialStudies,2,527-551.Carter,R.B.,andVanAuken,H.E.(1990)“SecurityAnalysisandPortfolioManagement:ASurveyandAnalysis”JournalofPortfolioManagement,Summer,81-85.Chamberlain,T.W.,C.S.Cheung,andC.C.Y.Chang(1991),“TheFridaytheThirteenthEffect,”QuarterlyJournalofBusinessandEconomics,30(2):111-117.Chan,L.,N.Jegadeesh,andJ.Lakonishok(1996),“MomentumStrategies,”NBERWorkingPaper5375,forthcoming,JournalofFinance.Fama,E.F.(1970),“EfficientCapitalMarkets:AReviewofEmpiricalWork,”JournalofFinance,25:383-417.Fama,E.F.(1991), “EfficientCapitalMarkets,”JournalofFinance,46(5):1575-1617. Fama,E.F.(1992),“TheCross-SectionofExpectedStockReturns,”JournalofFinance,47:427-465.Fama,E.F.(1997),"MarketEfficiency,Long-TermReturns,andBehavioralFinance,"CenterforResearchinSecurityPricesWorkingPaper448,UniversityofChicago.Fama,E.F.andBlume,M.(1966).“FilterRulesandStockMarketTradingProfits,”JournalofBusiness,39,SpecialSupplement,226-241King,W.I.(1934),TechnicalMethodsofForecastingStockPrices,JournaloftheAmericanStatisticalAssociation29,323-325.Jegadeesh,N.andS.Titman(1993),"ReturntoBuyingWinnersandSellingLosers:ImplicationsforStockMarketEfficiency,"JournalofFinance,48(1):65-91.Levich,R.M.(1985)“EmpiricalStudiesofExchangesRates:PriceBehaviour,RateDeterminationandMarketEfficiency,”inJones,R.andKenen,P.,eds,HandbookofInternationalEconomics,Amsterdam:North-HollandPublishingLevich,R.M.(1985),“IstheForeignExchangeMarketEfficient?”OxfordReviewofEconomicPolicy,Fall,40-60Levich,R.M.andL.R.Thomas(1993),Thesignificanceoftechnicaltrading-ruleprofitsintheforeignexchangemarket:abootstrapapproach,JournalofInternationalMoneyandFinance12,451-474.Lo,A.W.,H.Mamaysky,andJ.Wang(2000),FoundationsofTechnicalanalysis:ComputationalAlgorithms,StatisticalInference,andEmpiricalImplementation,JournalofFinance55,1705-1770.Lo,A.W.andA.C.MacKinlay(1990),WhenAreContrarianProfitsDuetoStockMarketOverreaction?"ReviewofFinancialStudies,3:175-208.Pring,M.(1997)“TechnicalAnalysisExplained”,NewYork:McGraw-HillThaler,R.H.(1987),"TheJanuaryEffect,"JournalofEconomicPerspectives,1(1): 197-201.Thaler,R.H.(1987),"SeasonalMovementsinSecurityPricesII:Weekend,Holiday,TurnoftheMonthandIntradayEffects,"JournalofEconomicPerspectives,1(1):169-177.Thorp,W.A.(2001),PredictingShort-TermTrends:theCup-With-HandlePattern,AAIIJournal(August),31-35.