CFA level 1conception.docx

CFA level 1conception.docx

《CFA level 1conception.docx》由会员分享，可在线阅读，更多相关《CFA level 1conception.docx（40页珍藏版）》请在冰豆网上搜索。

CFAlevel1conception

Thecurrencywiththehigher（lower）interestratewillalwaystradeatadiscount（premium）intheforwardmarket.Thelowerinterestrateinthedomesticcountrywillbeoffsetbytheappreciationofthedomesticcountry’scurrencyovertheinvestmenthorizon.

ThevalueofaEuropeanputoptionwilldecreaseastherisk-freeinterestrateincreases.

Whenacommoditymarketisincontango,futurespricesarehigherthanspotprices.Whenspotpricesarehigherthanthefuturesprice,themarketissaidtobeinbackwardation.

Bisincorrectbecausebackwardationistheoppositeofcontango;thefuturespriceisbelowthespotprice.

Cisincorrectbecausecarryreferstostorageplusinterestcosts.Itdoesnotsayanythingaboutfuturespricesrelativetospotprices.

Adistributionthatismorepeakedthannormaliscalledleptokurtic.

Adistributionthatisneithermorepeakednorlesspeakedthannormaliscalledmesokurtic.

Adistributionthatislesspeakedthannormaliscalledplatykurtic.

Thedurationofaportfolioistheweightedaverageofthebonds’durationsinwhichtheweightforeachbondisitscontributiontotheportfolio'svalue

MonteCarlosimulationlendsitselfto“whatif”analysisandrequirestheusertoprovideaprobabilitydistributionordistributions.Itcanbeacomplementtoanalyticalmethods.

MonteCarlosimulationprovidesadistributionofpossiblesolutionstocomplexfunctions.Thecentraltendencyandthevarianceofthedistributionofsolutionsgiveimportantcluestodecisionmakersregardingexpectedresultsandrisk.

Thesamplingerroristhedifferencebetweentheobservedvalueofastatisticandthequantityitisintendedtoestimate.

ZZ-score（standardizedvalue）=（X −μ）/σ，计算取值X距离平均值μ的标准差的个数

Themostappropriatetestischi-square,with36−1=35degreesoffreedom.Reminber:

Thedominatorjustforcalculatingthestandarddeviationofsamplewithsizeofnis（n-1）.

Anormaldistributionhaskurtosisof3.0.

whileforthecalculatingoftstatistic,thestandarderrorofsampleinformulaiss/n^0.5.

Thestandarderrorofthesample=StandardDeviation/n^0.5

ThesecuritymarketlineisagraphicalrepresentationoftheCAPMwithbetaonthe x-axisandexpectedreturnonthe y-axis.Theslopeofthelineisgivenbythemarketriskpremium,thedifferencebetweentheequitymarketreturnandtherisk-freerateofinterest.

ThebetaofStockA=CorrelationofStockAwiththemarket×StandarddeviationofStockA’sreturn÷Standarddeviationofthemarket’sreturn

Theoptimalriskyportfolioliesatthepointoftangencybetweenthecapitalallocationlineandtheefficientfrontierofriskyassets.

 theoptimalinvestorportfolioliesatthepointoftangencybetweentheinvestor’sindifferencecurveandthecapitalallocationline.

theglobalminimum-varianceportfolioistheleft-mostpointontheminimum-variancefrontier.

Ifaclientoffersabonusthatdependonthefutureperformanceofheraccount,thisisanadditionalcompensationarrangementthatrequireswrittenconsentinadvance.Ifaclientoffersabonustorewardamemberforheraccount’spastperformance,thisisagiftthatrequiresdisclosuretothemember’semployerwithstandardI（B）IndependenceandObjective.

 Foraninvestorwhoholdsafullydiversifiedportfolio,theTreynorratioandJensen’salphaaretheappropriateportfolioperformancemeasures.Theyareappropriatebecauseinafullydiversifiedportfolio,onlysystematicriskmatters;boththesemetricsmeasureperformancerelativetobetaorsystematicrisk.

TheTreynorratiomeasuresthereturnpremiumofaportfolioversustherisk-freeassetrelativetotheportfolio’sbeta,whichisameasureofsystematicrisk.

TheSharperatioissimilartotheTreynorratio,butitusesportfoliostandarddeviation,whichisameasureoftotalrisk,insteadofstandarddeviation.

M-squaredincorporatesthestandarddeviationofthemarketandportfolio,whicharemeasuresoftotalrisk.

TheCAPMrequiresthattherearenorestrictionsonshortselling（whichisanassumptionunderlyingfrictionlessmarkets）andthattheamountinvestedinanassetcanbeasmuchoraslittleastheinvestorwants（thatis,investmentsareinfinitelydivisible）.TheCAPMalsoassumesthatallinvestorsanalyzesecuritiesinthesamewayusingthesameinputsforfuturecashflowsandthesameprobabilitydistributions;thatis,itassumesthatinvestorshavehomogenousexpectations.

TheprovisionswithintheGIPSstandardsaredividedintothefollowingninesections:

FundamentalsofCompliance,InputData,CalculationMethodology,CompositeConstruction,Disclosure,PresentationandReporting,RealEstate,PrivateEquity,andWrapFee/SeparatelyManagedAccount（SMA）Portfolios.

TheinvestmentdecisionruleusingIRR（InternalRateofReturn）,theIRRrule,statesthefollowing:

“AcceptprojectsorinvestmentsforwhichtheIRRisgreaterthantheopportunitycostofcapital.”TheIRRruleusestheopportunitycostofcapitalasahurdlerate,orratethataproject’sIRRmustexceedfortheprojecttobeaccepted.NotethatiftheopportunitycostofcapitalisequaltotheIRR,thentheNPVisequalto0.Iftheproject’sopportunitycostislessthantheIRR,theNPVisgreaterthan0（usingadiscountratelessthantheIRRwillmaketheNPVpositive）.

Thetime-weightedrateofreturnmeasuresthecompoundrateofgrowthof$1initiallyinvestedintheportfoliooverastatedmeasurementperiod.Incontrasttothemoney-weightedrateofreturn,thetime-weightedrateofreturnisnotaffectedbycashwithdrawalsoradditionstotheportfolio.Theterm“time-weighted”referstothefactthatreturnsareaveragedovertime.Tocomputeanexacttime-weightedrateofreturnonaportfolio,takethefollowingthreesteps:

1.Pricetheportfolioimmediatelypriortoanysignificantadditionorwithdrawaloffunds.Breaktheoverallevaluationperiodintosubperiodsbasedonthedatesofcashinflowsandoutflows.

2.Calculatetheholdingperiodreturnontheportfolioforeachsubperiod.

3.Linkorcompoundholdingperiodreturnstoobtainanannualrateofreturnfortheyear（thetime-weightedrateofreturnfortheyear）.Iftheinvestmentisformorethanoneyear,takethegeometricmeanoftheannualreturnstoobtainthetime-weightedrateofreturnoverthatmeasurementperiod.

Themoneymarketisthemarketforshort-termdebtinstruments（one-yearmaturityorless）.Someinstrumentsrequiretheissuertorepaythelendertheamountborrowedplusinterest.Othersarepurediscountinstrumentsthatpayinterestasthedifferencebetweentheamountborrowedandtheamountpaidback.

Alldatameasurementsaretakenononeoffourmajorscales:

nominal,ordinal,interval,orratio.

1.Nominalscalesrepresenttheweakestlevelofmeasurement:

Theycategorizedatabutdonotrankthem.Ifweassignedintegerstomutualfundsthatfollowdifferentinvestmentstrategies,thenumber1mightrefertoasmall-capvaluefund,thenumber2toalarge-capvaluefund,andsoonforeachpossiblestyle.Thisnominalscalecategorizesthefundsaccordingtotheirstylebutdoesnotrankthem.

2.Ordinalscalesreflectastrongerlevelofmeasurement.Ordinalscalessortdataintocategoriesthatareorderedwithrespecttosomecharacteristic.Forexample,theMorningstarandStandard&Poor’sstarratingsformutualfundsrepresentanordinalscaleinwhichonestarrepresentsagroupoffundsjudgedtohavehadrelativelytheworstperformance,withtwo,three,four,andfivestarsrepresentinggroupswithincreasinglybetterperformance,asevaluatedbythoseservices.Anordinalscalemayalsoinvolvenumberstoidentifycategories.Forexample,inrankingbalancedmutualfundsbasedontheirfive-yearcumulativereturn,wemightassignthenumber1tothetop10percentoffunds,andsoon,sothatthenumber10representsthebottom10percentoffunds.Theordinalscaleisstrongerthanthenominalscalebecauseitrevealsthatafundranked1performedbetterthanafundranked2.Thescaletellsusnothing,however,aboutthedifferenceinperformancebetweenfundsranked1and2comparedwiththedifferenceinperformancebetweenfundsranked3and4,or9and10.

3.Intervalscalesprovidenotonlyrankingbutalsoassurancethatthedifferencesbetweenscalevaluesareequal.Asaresult,scalevaluescanbeaddedandsubtractedmeaningfully.TheCelsiusandFahrenheitscalesareintervalmeasurementscales.Thedifferenceintemperaturebetween10°Cand11°Cisthesameamountasthedifferencebetween40°Cand41°C.Wecanstateaccuratelythat12°C=9°C+3°C,forexample.Nevertheless,thezeropointofanintervalscaledoesnotreflectcompleteabsenceofwhatisbeingmeasured;itisnotatruezeropointornaturalzero.ZerodegreesCelsiuscorrespondstothefreezingpointofwater,nottheabsenceoftemperature.Asaconsequenceoftheabsenceofatruezeropoint,wecannotmeaningfullyformratiosonintervalscales.Asanexample,50°C,althoughfivetimesaslargeanumberas10°C,doesnotrepresentfivetimesasmuchtemperature.Also,questionnairescalesareoftentreatedasintervalscales.Ifaninvestorisaskedtorankhisriskaversiononascalefrom1（extremelyrisk-averse）to7（extremelyrisk-loving）,thedifferencebetweenaresponseof1andaresponseof2issometimesassumedtorepresentthesamedifferenceinriskaversionasthedifferencebetweenaresponseof6andaresponseof7.Whenthatassumptioncanbejustified,thedataaremeasuredonanintervalscale.

4.Ratioscalesrepresentthestrongestlevelofmeasurement.Theyhaveallthecharacteristicsofintervalmeasurementscalesaswellasatruezeropointastheorigin.Withratio