R5TimeValueR5TimeValueofDecomposerequiredrateofNominalrisk-freerate=realrisk-freerate+expectedinflationRequiredinterestrateona=nominalrisk-freerate+defaultriskpremium+liquidityriskpremium+maturityriskpremium考察方法Realrisk-freerate和nominalrisk-freerate3-R5TimeValueofEARrm R5TimeValueofEARrm 1EARm那么如果是semi,m=2;如果是quarterly,annual如果是連續(xù)復(fù)利，公式則變?yōu)榭疾旆椒?計(jì)算——算EAR，或者是算計(jì)息次定性（EAR和計(jì)息次數(shù)有關(guān)ThegreaterthecompoundingthegreatertheEARwillbeincomparisontothestatedthegreaterthedifferencebetweenEARandthestated4-R5TimeValueofTypeofOrdinaryannuity（后付年金AnnuityR5TimeValueofTypeofOrdinaryannuity（后付年金Annuitydue（先付年金Definition:anannuitywheretheannuitypaymentsoccuratthebeginningofeachcompoundingperiod.Measure1:putthecalculatorintheBGNmodeandinputrelevantMeasure2:treatasanordinaryannuityandsimplymultipletheresultingPVby(1+I/Y)（永續(xù)年金Definition:Aperpetuityisafinancialinstrumentsthatpaysafixedamountofmoneyatsetintervalsoveraninfiniteperiodoftime.PV=PMT+ 5-R6DiscountedCashFlowDiscountedR6DiscountedCashFlowDiscountedCashFlow NPV& Money-weightedreturn&Time-weighted6-R6DiscountedCashFlowNt12NNt 012t(1r(1r(1R6DiscountedCashFlowNt12NNt 012t(1r(1r(1r(1rNt tNPV0CF0...t IRR（InternalRateofWhenNPV=0,thediscountMultiplesolutionsProblemoftheIRRcalculation(#signBasicassumption:Reinvestmentrate=7-R6DiscountedCashFlowR6DiscountedCashFlowProjectDecisionSingleprojectNPVmethod:AcceptitifIRRmethod:AcceptitifIRR>r(requiredrateofTwoProjectsCaseSimilartoSingleprojectscaseMutuallyExclusiveProjectsNPVmethod:ChoosetheonewithhigherIRRmethod:ChoosetheonewithhigherNPVandIRRmethodsmayconflictwitheach8-R6DiscountedCashFlowDefine:theR6DiscountedCashFlowDefine:theholdingperiodreturnissimplythepercentagechangeinthevalueaninvestmentovertheperioditisHPR1P09-R6DiscountedCashFlow(FP0)360rBDFtR6DiscountedCashFlow(FP0)360rBDFtrtHPYP1P0EAY(1HPY)365/(1+BEY)2210-AninvestorbuysaAninvestorbuysaT-billat98,000with50daystomaturity.TheparvalueofthisT-billis100,000.Themoneymarketyieldisclosestto: Correctanswer:AU.S.Treasurybill(T-bill)has90daystomaturityandabankdiscountyieldof3.25%.Theeffectiveannualyield(EAY)fortheT-billisclosestto: Correctanswer:11-R6DiscountedCashFlowMoney-weightedandtime-weightedRateR6DiscountedCashFlowMoney-weightedandtime-weightedRateoftime-return掌握概念及公式概念：Time-weightedrateofreturnmeasurescompound步驟及公式：Firstly,computetheHPR;then,compute(1+HPR)forsubperiodtoobtainatotalreturnfortheentiremeasurementperiod[eg.(1+HPR1)*(1+HPR2)…(1+HPRn)].money-return掌握概念及公式概念：theIRRbasedonthecashflowsrelatedtothe步驟及公式：Firstly,determinethetimingofeachcashflow;then,usingthecalculationtocomputeIRR,orusinggeometricmean.考察方法：1.計(jì)算；注意計(jì)算time-return時(shí)，如果不是年度的不用開方；2.性質(zhì)：用TWRR衡量基金經(jīng)理的投資業(yè)績(jī)12-Ananalystgatheredthefollowinginformation($millions)abouttheperformanceofaportfolio:Theportfolioannualtime-weightedrateAnanalystgatheredthefollowinginformation($millions)abouttheperformanceofaportfolio:Theportfolioannualtime-weightedrateofreturnisclosest Correctanswer:13-Quarter(Priortoinfloworoutflow)Cashinflow(outflow)AtBeginningofQuarterValueatQuarter1234R7StatisticalConceptsandMarketStatisticalR7StatisticalConceptsandMarketStatistical Typesofmeasurement Measuresofcentral Chebyshev’s CV&Sharp Skewness&14-R7StatisticalConceptsandMarketR7StatisticalConceptsandMarketTypesofmeasurementNominaldistinguishingtwodifferentthings,noorder,onlyhasexample:assigningthenumber1toamunicipalbondfund,thenumber2toacorporatebondfund.Ordinalscales(>,makingthingsinorder,butthedifferencearenotexample:therankingof1,000smallcapgrowthstocksbyperformancemaybedonebyassigningthenumber1tothe100bestperformingstocksIntervalscales(>,<,+,-subtractisexample:TemperatureRatioscales(>,<,+,-,*,/)withoriginalexample:money,ifyouhavezerodollars,youhavenopurchasingpower,butifyouhave$4.00,youhavetwiceasmuchpurchasingpowerasapersonwith$2.00.15-R7StatisticalConceptsandMarketNThearithmeticXXiR7StatisticalConceptsandMarketNThearithmeticXXi nnXWwiXi(w1X1w2X2wnXnNGNX1X2X3...XN(Xi)1/ThegeometricinTheharmonicXHn(1/Xiiharmonicmean<=geometricmean<=arithmetic16-R7StatisticalConceptsandMarketR7StatisticalConceptsandMarketQuartileThethirdquartile:75%,orthree-fourthsoftheobservationsfallbelowthatvalue.CalculationLy=(n+1)y/100,LyistheQuantitlesandmeasuresofcentraltendencyareknowncollectivelyasmeasuresoflocation.ThemedianofadistributionisleastlikelyequaltoSecondThirdFifth Correctanswer: 17-R7StatisticalConceptsandMarketRange=maximumvalue–minimumNR7StatisticalConceptsandMarketRange=maximumvalue–minimumNXMAD nN(Xi Nn(XiX n18-R7StatisticalConceptsandMarketChebyshev’sForanyR7StatisticalConceptsandMarketChebyshev’sForanysetofobservations(samplesorpopulation),theproportionofvaluesthatliewithinkstandarddeviationsofthemeanisatleast1–wherekisanyconstantgreaterthan對(duì)任何一組觀測(cè)值，個(gè)體落在均值周圍k個(gè)標(biāo)準(zhǔn)差之內(nèi)的概率不小于1/k2，對(duì)任意k>1Thisrelationshipappliesregardlessoftheshapeofthe19-R7StatisticalConceptsandMarketCoefficientR7StatisticalConceptsandMarketCoefficientofvariationmeasurestheamountofdispersioninarelativetothedistribution’smean.(relativeThesharpratiomeasuresexcessreturnperunitofSharpratio=RP-CV=sxX20-R7StatisticalConceptsandMarketPositive(right)Negative(left)Positiveskewed：Mode<median<mean,R7StatisticalConceptsandMarketPositive(right)Negative(left)Positiveskewed：Mode<median<mean,havingarightfatNegativeskewed：Mode>media>mean,havingaleftfatInvestorsshouldbeattractedbyapositiveskewbecausethemeanreturnfallsabovethemedian.Samplenn(XiX(XiX (n1)(n1SK] ()i1 n考察方法根據(jù)描述的特點(diǎn)判斷是Positivelyskewed還是Negative21-R7StatisticalConceptsandMarketLeptokurticvs.ItdealswithR7StatisticalConceptsandMarketLeptokurticvs.Itdealswithwhetherornotadistributionismoreorless“peaked”thananormaldistributionExcesskurtosis=samplekurtosis–考察方法根據(jù)描述的特點(diǎn)判斷是leptokurtic還是根據(jù)已知的峰度，選擇都有哪些特可能在考試中會(huì)和skew合并考核綜合知22-NormalSampleExcessR7StatisticalConceptsandMarketR7StatisticalConceptsandMarketFatAleptokurticreturndistributionhasmorefrequentextremelylargedeviationsfromthemeanthananormaldistribution.23-R8ProbabilityR8ProbabilityTwodefiningpropertiesofEmpirical,subjective,andprioriOddsfororMultiplicationruleandadditionDependentandindependentCovariance&Expectedvalue,variance,andstandarddeviationofarandomvariableandofreturnsonaportfolioBayes’24-R8ProbabilityJointprobabilityR8ProbabilityJointprobability:MultiplicationP(AB)=P(A|B)×P(B)=IfAandBaremutuallyexclusiveevents,then:ProbabilitythatatleastoneoftwoeventswillAdditionP(AorB)=P(A)+P(B)-IfAandBaremutuallyexclusiveevents,then:P(AorB)=P(A)+P(B)25-R8ProbabilityTheoccurrenceofAhasR8ProbabilityTheoccurrenceofAhasnoinfluenceofontheoccurrenceofP(A|B)=P(A)orP(AorB)=P(A)+P(B)-IndependenceandMutuallyExclusivearequiteIfexclusive,mustnotCauseexclusivemeansifAoccur,Bcannotoccur,Ainfluents26-R8ProbabilityForunconditionalprobabilityofeventP(A)R8ProbabilityForunconditionalprobabilityofeventP(A)P(AS1)P(S1)P(AS2)P(S2)...P(ASN)P(S)wherethesetofeventsS1,S2,...SNismutuallyexclusiveE(X)P(Xi)XiExpectedE(X)*P(xi)x1*P(x1)x2*P(x2)N*P(xnP(EX 2ii27-R8ProbabilityCovariancemeasureshowonerandomvariablemoveswithanotherR8ProbabilityCovariancemeasureshowonerandomvariablemoveswithanotherrandomThecovarianceofRAwithitselfisequaltothevarianceofCovariancerangesfromnegativeinfinitytopositiveCOV(X,X)E[(X-E(X))(X-E(X))]2COV(X,Y)E[(X-E(X))(Y-COV(X,Y) CorrelationmeasuresthelinearrelationshipbetweentworandomCorrelationhasnounits,rangesfrom–1to+1,standardizationofUnderstandthedifferencebetweencorrelationandIfρ=0,this28-AnanalystgatheredinformationAnanalystgatheredinformationaboutthreeeconomicvariables,HenotedthatwhenevervariableAincreasedbyoneunit,variableBincreasedby0.5unitsbutvariableCdecreasedby0.5units.ThecorrelationbetweenvariablesAandBandthecorrelationbetweenvariablesAandCrespectively,areclosestto:CorrelationbetweenvariablesAandBAnswerAnswerAnswerCCorrectanswer:CCorrelationbetweenvariablesAand---A29-R8ProbabilityTheexpectedvalue,variance,andstandardR8ProbabilityTheexpectedvalue,variance,andstandarddeviationofarandomvariableandofreturnsonaportfolio;nE(rp)wiE(Ri wwcov(R,R2p iji130-R8ProbabilityCalculateandinterpretanupdatedprobabilityusingBayes’R8ProbabilityCalculateandinterpretanupdatedprobabilityusingBayes’P(AB)=P(A|B)×P(B)P(B|A)*P(AP(A|B)P(BP(R|Si)P(SiP( |R)iP(R31-R9CommonProbabilityR9CommonProbabilityCommonProbabilityPropertiesofdiscretedistributionandcontinuousUniformrandomvariableandabinomialrandomThekeypropertiesofthenormalStandardizearandomConfidenceintervalforanormallydistributedrandomLognormalSafety-firstMonteCarlo32-R9CommonProbabilityProbabilityR9CommonProbabilityProbabilityDescribetheprobabilitiesofallthepossibleoutcomesforarandomDiscreteandcontinuousrandomDiscreterandomvariables:thenumberofpossibleoutcomescanbecounted,andforeachpossibleoutcome,thereisameasurableandpositiveprobability.Continuousvariables:thenumberofpossibleoutcomesisinfinite,eveniflowerandupperboundsexist.P(x)=0eventhoughxcanP33-R9CommonProbabilityProbabilityR9CommonProbabilityProbabilityFordiscreterandom0≤p(x)≤Probabilitydensityfunction(p.d.f):ForcontinuousrandomvariableCumulativeprobabilityfunction(c.p.f):34-R9CommonProbabilityBinomialBernoullirandomP(Y=0)=1-BinomialrandomvariableR9CommonProbabilityBinomialBernoullirandomP(Y=0)=1-Binomialrandomvariable，theprobabilityofxsuccessesinnp)np(x)P(XExpectationsandx)xx35-Bernoullirandomvariablepp(1-Binomialrandomvariablenp(1-R9CommonProbabilityContinuousUniformR9CommonProbabilityContinuousUniform--isdefinedoverarangethatspansbetweensomelowerlimit,a,andlimit,b,whichserveastheparametersofthePropertiesofContinuousuniformForalla≤x1<x2P(x1Xx2)(xP(X<aorX>b)=x1)/(ba36-R9CommonProbabilityTheshapeR9CommonProbabilityTheshapeofthedensityxX~N(μ,Symmetricaldistribution:skewness=0;Alinearcombinationofnormallydistributedrandomvariablesisalsonormallydistributed.Thetailsgetthinandgotozerobutextendinfinitely,asympotic(漸近37-R9CommonProbabilityTheconfidence68%confidenceinterval90%confidenceinterval95%confidenceinterval99%confidenceR9CommonProbabilityTheconfidence68%confidenceinterval90%confidenceinterval95%confidenceinterval99%confidenceinterval[1.65,1.65][1.96,1.96][2.58,2.58][,U- 38-R9CommonProbabilityStandardnormalN(0,1)R9CommonProbabilityStandardnormalN(0,1)orStandardization:ifX~N(μ,σ2),thenZX~Z-F(-z)=1-P(Z>z)=139-R9CommonProbabilityShortfallrisk:RL=R9CommonProbabilityShortfallrisk:RL=thresholdlevelreturn,minimumreturnMinimize(Rp<RL)[E(RP)RL]/PMaximizeS-F-MaximizeSFR=E(RP)-RLMinimizeP(Rp<40-R9CommonProbabilityDefinition:R9CommonProbabilityDefinition:IflnXisnormal,thenXislognormal,whichisusedtodescribethepriceofassetRightBoundedfrombelowbyThelognormaldistributionisusedtomodelasset41- 1R9CommonProbabilityMonteCarlosimulationvsHistoricalMonteCarlosimulationR9CommonProbabilityMonteCarlosimulationvsHistoricalMonteCarlosimulationusesrandomlygeneratedvaluesforriskfactors,basedontheirassumeddistributions,toproduceadistributionofpossiblesecurityvalues,toanalyzethecomplexinstrument;ItisfairlycomplexandwillassumeaparameterItisnotananalyticmethodbutastatisticalone,andcannotprovidetheinsightsthatanalyticmethodscan.Historicalsimulationusesrandomlyselectedpastchangesintheseriskfactorstogenerateadistributionofpossiblesecurityvalues,can’tanswerthe“What-If”.Limitations:thepastcannotindicatethefutureandhistoricalsimulationaddressthesortof“whatif”questionsthatMonteCarlosimulation42-R10SamplingandR10SamplingandSamplingandSimplerandomandstratifiedrandomsampling,time-seriesandcross-sectionaldataCentrallimitStandarderrorofthesamplemeanThedesirablepropertiesofanStudent’st-distributionCriteriaforselectingtheappropriateteststatistic，計(jì)算confidenceFivekindsof43-R10SamplingandSamplingR10SamplingandSamplingandSimplerandomStratifiedrandomsampling:toseparatethepopulationintosmallerbasedononeormoredistinguishingcharacteristics.StratumandSamplingerror:samplingerrorofthemean=samplemean-populationThesamplestatisticitselfisarandomvariableandhasaprobability44-R10SamplingandTime-seriesconsistofR10SamplingandTime-seriesconsistofobservationstakenoveraperiodoftimeatspecificandequallyspacedtimeintervals.Cross-sectionalasampleofobservationstakenatasinglepointin45-Time-seriesCross-sectionalacollectionofdatarecordedoveraperiodoftimeacollectionofdatatakenatasinglepointoftime.R10SamplingandCentralLimitForsimplerandomsamplesofsizenfromapopulationwithameanμandavarianceσ2butwithoutknowndistribution,thesamplingdistributionofthesamplemeanapproachesN(μ,σ2/n)R10SamplingandCentralLimitForsimplerandomsamplesofsizenfromapopulationwithameanμandavarianceσ2butwithoutknowndistribution,thesamplingdistributionofthesamplemeanapproachesN(μ,σ2/n)ifthesamplesizeissufficientlylarge(nn條件:2.總體均值方差都存結(jié)論1.服從正態(tài)分nStandarderrorofthesampleKnownpopulationnnxUnknownpopulation s46-R10SamplingandTheR10SamplingandThedesirablepropertiesofanUnbiasedness:expectedvalueoftheestimatorisequaltotheparameterthataretryingtoestimateEfficiency:forallunbiasedestimators,ifthesamplingdispersionissmallerthananyotherunbiasedestimators,thenthisunbiasedestimatoriscalledefficient.Consistency:theaccuracyoftheparameterestimateincreasesasthesamplesizeincreases.(thestandarddeviationoftheparameterestimatedecreasesasthesamplesizeincreases)Asthesamplesizeincreases,thestandarderrorofthesamplemean47-R10SamplingandR10SamplingandPointestimate:thestatistic,computedfromsampleinformation,whichisusedtoestimatethepopulationparameterConfidenceintervalestimate:arangeofvaluesconstructedfromsampledatasotheparameteroccurswithinthatrangeataspecifiedprobability.α—thelevelofsignificanceIntervalEstimation（alsoseeChapter:HypothesisTestingLevelofsignificanceDegreeofConfidenceConfidenceInterval=[PointEstimate+/-(reliabilityfactor)*Standard48-R10SamplingandDegreesoffreedom(df):n-R10SamplingandDegreesoffreedom(df):n-Lesspeakedthananormaldistribution(“fatterAsthedegreesoffreedomgetslarger,theshapeoft-distributionstandardnormal012349-R10SamplingandCalculateandinterpretaconfidenceintervalforapopulationmean,givenanormaldistributionwith1)aknownpopulationvariance,2)anunknownpopulationvariance,or3)anunknownvarianceandalargesamplesize;sxR10SamplingandCalculateandinterpretaconfidenceintervalforapopulationmean,givenanormaldistributionwith1)aknownpopulationvariance,2)anunknownpopulationvariance,or3)anunknownvarianceandalargesamplesize;sxx22nnWhensamplingformTestsamplesampleNormaldistributionwithknownvarianceNormaldistributionwithunknownvarianceNonnormaldistributionwithknownvarianceNonnormaldistributionwithunknownt-notnotz-t-z-Statistic50-R11HypothesisR11HypothesisHypothesisThestepsofhypothesisThenullhypothesisandalternativehypothesis,one-tailedandtwo-tailedtestTypeIandtypeIIDecisionTheChi-squaretestandF-Parametertestsandnon-parameter51-R11HypothesisStepsofhypothesisStepStateR11HypothesisStepsofhypothesisStepStatenullandStepIdentifythetestStepSelectalevelofStepStepDonotarriveatFormulateadecision52-R11HypothesisDefineNullhypothesisandAlternativehypothesis(wewanttoR11HypothesisDefineNullhypothesisandAlternativehypothesis(wewanttoH0:Ha:One-tailedandTwo-tailedtestsof : :0aH0:Ha:or, : :00a053-R11HypothesisTestTestforpopulationR11HypothesisTestTestforpopulation s Criticalvalue(關(guān)鍵值，實(shí)際就是分位數(shù)FoundintheZ,T,ChiSquareorFdistributiontablesnotcalculatedbyUndergivenonetailedortwotailedassumption,criticalvalueissolelybythesignificance54-R11Hypothesis-FailtoFailtoRejectRejectR11Hypothesis-FailtoFailtoRejectRejectRejectRejectH0if|teststatistic|>criticalFailtorejectH0if|teststatistic|<criticalcannotsay“acceptthenullhypothesis”,onlycansay“cannot*****issignificantlydifferentfrom*****isnotsignificantlydifferentfrom55-R11SummaryofHypothesisCriticalZxhypothesisknownpopulation xNormallydistributedpopulation,unknownpopulationvariancet0nsIndependentpopulations,assumedequaltt(n1+n2Independentpopulations,R11SummaryofHypothesisCriticalZxhypothesisknownpopulation xNormallydistributedpopulation,unknownpopulationvariancet0nsIndependentpopulations,assumedequaltt(n1+n2Independentpopulations,notassumedequalttdsSamplesnotpairedcomparisonstd(n1)sNormallydistributed2(n20FF(n11,n2distributedpopulations1256-R11HypothesisP-valueR11HypothesisP-valueThep-valueisthesmallestlevelofsignificanceatwhichthenullhypothesiscanberejectp-value<α:rejectH0;p-value>α:donotrejectP↓,easiertorejectH0Atwo-tailedtestofthenullhypothesisthatthemeanofadistributionisequalhasap-valueof0.0567.Usinga5%levelofsignificance(i.e.,α=0.05),thebestconclusionisto:rejectthenullacceptthenullincreasethelevelofsignificanceto5.67%.Correctanswer:B57-R11HypothesisTrueisisDonotR11HypothesisTrueisisDonotCorrectIncorrectTypeⅡIncorrectCorrectSignificancelevelPoweroftest1-P(TypeⅡH0=P(TypeIWithotherconditionsunchanged,eithererrorprobabilityarisesatthecostoftheothererrorprobabilitydecreasing.Howtoreducebotherrors?IncreasetheSample58-Ifatwo-tailedhypothesisIfatwo-tailedhypothesistesthasa5%probabilityofrejectingthenullhypothesiswhenthenullistrue,itismostlikelythatthe:PowerofthetestisConfidencelevelofthetestisProbabilityofaTypeIerroris2.5%Correctanswer:BWhichofthefollowingstatementsabouthypothesistestingismostRejectingatruenullhypothesisisaTypeIThepowerofatestistheprobabilityoffailingtorejectthenullhypothesiswhenitisfalse.Foraone-tailedtestinvolvingX,thenullhypothesiswouldbeH0:X=0,andthealternativehypothesiswouldbeHA:X≠0.Correctanswer:59-JoeBay,CFA,wantsJoeBay,CFA,wantstotestthehypothesisthatthevarianceofreturnsonenergystocksisequaltothevarianceofreturnsontransportationstocks.Bayassumesthesamplesareindependentandthereturnsarenormallydistributed.Theappropriateteststatisticforthishypothesisisa(n):Correctanswer:BAliceMorton,CFA,isreviewingaresearchpaperthatreachesaconclusionbasedontwohypothesiswithp-valuesof0.037and0.064.Mortonshouldconcludethat:Bothofthesetests’nullhypothesescanberejectedwith90%Neitherofthesetests’nullhypothesescanberejectedwith95%Onlyoneofthesetests’nullhypothesescanberejectedwith99%confidence.Correctanswer:A60-Usingthesampleresultsgivenbelow,drawnas25pairedobservationsfromtheirunderlyingdistributions,testifthemeanreturnsofthetwoportfoliosdifferfromeachotheratthe1%levelofstatisticalsignificance.AssumetheunderlyingdistributionsofreturnsforeachportfolioarenormalandthattheirpopulationUsingthesampleresultsgivenbelow,drawnas25pairedobservationsfromtheirunderlyingdistributions,testifthemeanreturnsofthetwoportfoliosdifferfromeachotheratthe1%levelofstatisticalsignificance.Assumetheunderlyingdistributionsofreturnsforeachportfolioarenormalandthattheirpopulationvariancesarenotknown.Basedonthepairedcomparisonstestofthetwoportfolios,themostappropriateconclusionis:rejectthehypothesisthatthemeandifferenceequalszeroasthecomputedteststatisticexceeds2.807.acceptthehypothesisthatthemeandifferenceequalszeroasthecomputedteststatisticexceeds2.807.acceptthehypothesisthatthemeandifferenceequalszeroasthecomputedteststatisticislessthan2.807.Correctanswer:61-PortfolioPortfolioMeant‐statisticfor24dfandatthe1%levelofstatisticalsignificance=R12TechnicalR12TechnicalTechnicaltheprinciplesoftechnicalanalysis,itsapplications,anditsunderlyingTypesoftheusesofCommonchartCommonanalysistheuseof62-R12TechnicalPricesaredeterminedbytheinteractionofR12TechnicalPricesaredeterminedbytheinteractionofsupplyandOnlyparticipantswhoactuallytradeaffectprices,andbetter-informedparticipantstendtotradeingreatervolume.Priceandvolumereflectthecollectivebehaviorofbuyersandsellers.MarketpricesreflectbothrationalandirrationalinvestorInvestorbehaviorisreflectedintrendsandpatternsthattrendtorepeatandcanbeidentifiedandusedforforecastingprices.Efficientmarketshypothesisdosenot63-R12TechnicalThedifferencesamongtechnicians,fundamentalistsandEfficientFundamentalanalysisofafirmattemptstoR12TechnicalThedifferencesamongtechnicians,fundamentalistsandEfficientFundamentalanalysisofafirmattemptstodeterminetheintrinsicvalueanassetbyusingthefinancialstatementsandotherTechnicalanalysisusesonlythefirm’ssharepriceandtradingvolumedata,anditisnotconcernedwithidentifyingbuyers’andsellers’reasonsfortrading,butonlywiththetradesthathaveFundamentalistsbelievethatpricesreactquicklytochangingstockwhiletechniciansbelievethatthereactionisslow.Technicianslookchangesinsupplyanddemand,whilefundamentalistslookforchanges64-R12TechnicalAdvantagesR12TechnicalAdvantagesoftechnicalActualpriceandvolumedataareTechnicalanalysisitselfisobjective(althoughrequiresubjectivejudgment),whilemuchofthedatausedinfundamentalanalysisissubjecttoassumptionsorrestatements.Itcanbeappliedtothepricesofassetsthatdonotproducefuturecashflows,suchascommodities.Itcanalsobeusefulwhenfinancialstatementfraudoccurs.Theusefulnessislimitedinmarketswherepriceandvolumedatamightnottrulyreflectsupplyanddemand,suchasinilliquidmarketsandinmarketsthataresubjecttooutside65-CFA一級(jí)培訓(xùn)項(xiàng)CFA一級(jí)培訓(xùn)項(xiàng)PortfolioTopicWeightingsinCFALevel2-StudySessionEthics&ProfessionalStudySession2-QuantitativeTopicWeightingsinCFALevel2-StudySessionEthics&ProfessionalStudySession2-QuantitativeStudySession4-StudySession7-FinancialReportingandStudySessionCorporate7StudySessionPortfolio7StudySession13-EquityStudySession15-FixedStudySession5StudySessionAlternative4PortfolioRiskandReturn:PartAnindividualExpectednE(R)PortfolioRiskandReturn:PartAnindividualExpectednE(R)iPRPPn1 nVar=2[RiE(R)]2VarianceofnSD=[RE(R)]2StandardDeviationofii3-PortfolioRiskandReturn:PartPortfolioRiskandReturn:Part4-PortfolioRiskandReturn:PartTheportfoliostandarddeviationPortfolioRiskandReturn:PartTheportfoliostandarddeviationTheriskofaportfolioofriskyassetsdependsontheassetweightsandstandarddeviationsoftheassetsreturns,andcruciallyonthecorrelation(covariance)oftheassetreturns.Thelowerthecorrelationbetweenthereturnsofthestocksintheallelseequal,thegreaterthediversificationTwo-asset5-PortfolioRiskandReturn:PartRiskPortfolioRiskandReturn:PartRiskandreturnfordifferentvaluesof6-PortfolioRiskandReturn:PartThePortfolioRiskandReturn:PartTheminimum-varianceandefficientfrontiersofriskyassetsandtheglobalminimum-varianceportfolio.GlobalVarianceEfficient(AllEfficientInefficientIndividual7-PortfolioRiskandReturn:PartPortfolioRiskandReturn:PartRiskReferstothefactthatindividualspreferlessrisktomoreRisk-aversePreferlowertohigherriskforagivenlevelofexpectedWillonlyacceptariskierinvestmentiftheyarecompensatedintheformofgreaterexpectedreturn8-PortfolioRiskandReturn:PartTheoptimalportfolioPortfolioRiskandReturn:PartTheoptimalportfolioforanAtthepointofwhereaninvestor’s(highest)risk-returnindifferencecurveistangenttotheefficientfrontier.GlobalVarianceEfficient(AllEfficientInefficientIndividualOptimalThehighestindifferencecurvethatistangenttotheefficientDifferentinvestorsmayhavedifferentoptimal9-PortfolioRiskandReturn:PartTheimplicationsofcombiningarisk-freeassetwithaportfolioofriskyE(RPE(RPortfolioRiskandReturn:PartTheimplicationsofcombiningarisk-freeassetwithaportfolioofriskyE(RPE(RA)yWA0AE(RP)WAE(RA)WBE(RBWW2W P WW P 10-PortfoliowithWAInvestedintheRisPortfolioRiskandReturn:PartThePortfolioRiskandReturn:PartThecapitalallocationline(CAL)andthecapitalmarketlineTwo-fundseparationCombiningariskyportfoliowitharisk-freeAllinvestors’optimumportfolioswillbemadeupofsomecombinationanoptimalportfolioofriskyassetsandtherisk-freeThelinerepresentingthesepossiblecombinationsofrisk-freeassetsandoptimalriskyasset11-PortfolioRiskandReturn:PartRiskyPortfoliosandTheirPortfolioRiskandReturn:PartRiskyPortfoliosandTheirAssociatedCapitalAllocationLinesforDifferentE(RCALACALBCIfeachinvestorhasdifferentexpectationsabouttheexpectedreturnsof,standarddeviationsof,orcorrelationsbetweenriskyassetreturns,eachinvestorwillhaveadifferentoptimalriskyassetportfolioandadifferent12-PortfolioRiskandReturn:PartPortfolioRiskandReturn:PartTheMarketIsthetangentpointwheretheCMLtouchestheMarkowitzefficientConsistsofeveryriskyTheweightsoneachassetareequaltothepercentageofthemarketvaluetheassettothemarketvalueoftheentiremarket13-PortfolioRiskandReturn:PartCapitalmarketWheninvestorssharePortfolioRiskandReturn:PartCapitalmarketWheninvestorsshareidenticalexpectationsaboutthemeanreturns,varianceofreturns,andcorrelationsofriskyassets,theCALforallinvestorsisthesameandisknownasthecapitalmarketlineE(RM)E(R)PFPMExplanationoftheInvestmentusingCMLfollowapassiveinvestmentstrategy(i.e.,investinanindexofriskyassetsthatservesasaproxyforthemarketportfolioandallocateaportionoftheirinvestableassetstoarisk-freeasset.)BorrowingportfolioandlendingDifferencebetweentheCMLandthe14-PortfolioRiskandReturn:PartUnsystematicrisk(orunique,diversifiable,PortfolioRiskandReturn:PartUnsystematicrisk(orunique,diversifiable,firm-specificTheriskthatdisappearsintheportfolioconstructionprocessSystematicrisk(ormarketrisk):TheriskthatisleftcannotbediversifiedTotalrisk=systematicrisk+unsystematicBeta:thesensitivityofanasset’sreturntothereturnonthemarketindexinthemarketmodel.( )ii215-PortfolioRiskandReturn:PartHowtojudgeifaPortfolioRiskandReturn:PartHowtojudgeifastockisproperlyAppropriatelyundervalued,Overvalued,Beta,SystematicTheEquationofSMLE(Ri)= +βi[E(R)- fMf16-PortfolioRiskandReturn:PartDifferencesbetweentheSMLandPortfolioRiskandReturn:PartDifferencesbetweentheSMLandthe17-MeasureofUsessystematicrisk(non-diversifiablerisk)(totalrisk)Toolusedtodeterminetheappropriateexpected(benchmark)returnsforToolusedtodeterminetheappropriateassetallocation(percentagesallocatedtotherisk-freeassetandtothemarketportfolio)fortheinvestorGraphofthecapitalassetpricingmodelGraphoftheefficientMarketriskMarketportfolioSharpePortfolioRiskandReturn:PartSharpeSharperatio=RM-squaredRPortfolioRiskandReturn:PartSharpeSharperatio=RM-squaredR)RRMPfMfPTreynorTreynormeasure=Jensen’sP(RPRf)P(RM)18-PortfolioManagement:AnCharacteristicsofdifferenttypesofRiskInvestmentLiquidityPortfolioManagement:AnCharacteristicsofdifferenttypesofRiskInvestmentLiquidityIncomeDependsonPayinterestDependsonLong—lifeShort—DependsonDBMutualDependsonDependson19-PortfolioManagement:AnPlanningAnalysisofPortfoli