《CFA二级前导班培训项目：框架介绍_道德+数量+经济+固收+组合.docx》由会员分享，可在线阅读，更多相关《CFA二级前导班培训项目：框架介绍_道德+数量+经济+固收+组合.docx（49页珍藏版）》请在课桌文档上搜索。
1、CFA二级培训项目讲师：Vincent竺5Mg*IX新虫理Vincent工作职称工金程教育密深培训师-教胃背景工英国埃塞克斯大学金融学硕士、通过CFA三锣、PMP（ProjectManageBentPTofessional项目管理专业认证）-工作IIfh曾任某外资银行总部项目经理，十二年的外企银行工作责历，积K了丰富的金触实战经验.现为金程敦曾CFA赛深培训讲I,担任CFAM目教学产品研发负责人.熟悉CFA考试重点，擅长授课包括职业伦理、经济学、固定收益、数St分折、组合管ML授课逻辑清断易If,结合实际案例深入浅出解择考点，备受学员欢迎.2-106服务客户t中国银行.中Si建设银行、肮州联分
2、银行、机州银行、国泰君安证券、苏州元禾控股等有理囹RrlBllSTopicWeightingsinCFALevelIIISessionNO*ContentWeightingsIStudySession1-2Ethics&ProfessionalStandardS10-15StudySession3QuantitativeMethods5-10StudySesston4EconomicsS-IOStudySession5-6FinancialReportingandAnalysis10-15StudySession7-8CorporateFinance5-10StudySession9-11Eq
3、uityValuationIO-15StudySession12-13FixedIncome10-15StudySession14Derivatives5-10StudySession15AlternativeInvestmentsS-10StudySession16-17PortfolioManagement10-153-106百亚医顺71B二级学习方法一级二Si各科目权比较科目一级权重二级权重Ethics&ProfessionalStandards1510-15QuantitativeMethods105-10Economics105-lFinancialStatementAnalysis
4、15IO-ISCorporateFinance10510EquityValuation610-15FixedIncome1110-15derivatives115-10AlternativeInvestments65-10PortfolioManagement610-154106互理囹新ItflEthics&ProfessionalStandards-tB三FrameworkASSlEthkalandProfessionalStandardsRlCcxleofEthicsandStandardsofProfessionalConduct/2primaryprinciples/Proceedin
5、gs/Panel/AMCvsCodeandstandards/CodeofethicsSlWfttuFrameworkSSlEthicalandProfessionalStandardsR2GuidanceforStandardsI-V11/1:Professionalism/11:Integrityofcapitalmarkets/III:Dutiestoclients/IV:Dutiestoemployers/V:Investmentanalysis,recommendations,andactions/VI:Conflictsofinterest7-106/V11:Responsibil
6、itiesasaCFAInstitutememberorCFAcandidate* Guidance for Standards IL ProfssionaHSm 卜吨宙IIKInKnowledgeofthlaw/了修与工作相关的注博，法ML准则和济会COdeandStanda通:/避守最严格法律:/不得被意参与il法Ir为：/育M向咨询上司/合燃.不需要向我附福门举H除参法律明文嵯定：/发现HiMl行为.自己不作为.初为忌嫌：Ind*pndnceandobjectivity/直分礼物.基金经号的客户（/）给的是小费.broker给基金经理或者上市公司给aru9tt都是要影响客脱独立性的不能收
7、除等是modestgrfb/基金公司（买方）俺用分析师（卖方）I/防火应停RI离投行部与其他所有离门.q试特别修与研亢部的1高：/Issuerpaidresearch11Ifj.也必须收fUfee.井进bit露：/差*费自己出.除葬IQ曼工具无法到达的情况.可接受目标公司的一般安K：Misrepresentation/没有及时改正的打字博误：/不能胡说八it（美于个人宠质、关于公司服务前困、模里结果没有交代清整、业缄度和日因PaCs,比如不可以ChEyPdmg、不色捐像不该拒保的收益、使用外品修命经理要披露，业HJbecchEark出界不备当/不能方4L必列恰当引用：Misconduct/欺在
8、盛日等不信行为：/喝酒a箫酒：/任何伪者0如9信.信Ir工作展任能力的行为逢反m&:OndUul/与工作无美的个人信仰.政治修向的，议.不违反mcoCdUG/善欺作IMk的个人破产可以免贵.档案期内要披露.三W-*IlflGuidanceforStandardsIIMaterialonpublicrnfbrmation大必须同时具3消息来淹可靠.以及对股价有重大影响（举例参见崖件）j2.Integrityofcapital/非公开，向:eZave对象披露不属于公开：marketslMos理论*j/做市商掌握了MM不可以停止做市.应该做消极对手方：/从事无风险套利交易时.若获得了MNI,除非公司
9、有能力证明1.流程和记录短范才可以继续交易.否则要停止,Marketmanipulatxc/两片形式（关位看动机）：敢布Iy消息、基于交晶扭曲置价：/以避税为目的的交易.不违反市场爆ttU/基于特定的交易策略.不通反市场操纵,/如果为了堵加流动性，期货交易所会员，书面协议.并对外事先披露，则不住违埋.等业色*!窗GuidanceforStandardsIII13.DutyToClnt-1.oyalty,prudncndcare/Fidgary需要履行extrjcare,higherstandard:/四美客户*individual,beneficiarymandate,investingPUb
10、he:/基于第合整体进行判Vh/Softdollar(softcommission)JK.brokerage是客户的M产必甯100%让X户直接受标/Directbrokerage的情况F仍然宵义务寻找bestexecution和bestprice:/VOGnggXy必务1进行.除尊批于性价比号虐.要向客户IfvotmgProxyPOhCles：FairdMing/Falr井TXqUah可以有PfemlUmlev，lservc，但必渠械露.同时不仿害IC他客户利益：保客户有公允的机会时投责推荐出反应：/聚取round-IOtbass.避免odd-lotdtstnbutions：/family-m
11、emberaccounts井非beneficiary要与真ItX户一祝同仁：SuitebiHty/基于客户的IPS(RRTTLLU)，至少每年更新，投责重大交更必缪先修改IPS:/0燎分散化.基于组合整体角度，虐、/宫户囚执己见的文口.如果对整体彭看不大可以售闷.彭大0牙他改IPS.客户不同意修IMpSIt从管理联户中IM育贯金交让客户自己管理；PerformanceprMntation/过去业续不能示在未来。I以达成I/业飨而报可以.但必争同以后续提供洋H信总：/IH合怛：E杈平均.终止的组合.所有相IU凤格怛令.相关支埼记承保存：Preservationofconfidentiality保
12、定t对NFSIi去X户、旗在宫户、俞格滔在N户I是否保密先必须专虐法0缴定.10-106有曲倒Hft“GuidanceforStandardsIV4DutyTbemployers1.oyalty/IndePendemPQCtke是指与雇主业务在内容匕时间匕精力上相竟争的业务.必须告知雇主性质、expectedduration以及COmPenSatiOn并且得到许可：开雇主前不拿雇主一针一线正式离职前不可以先开始拉原客户:/仅仅知道几个客户是可以的.但不可以背诵客户名单：Additionalcompensationarrangements各方书面同意；，告知CoEpensatiOll的性质、大概
13、金桢、duration：Responsibilityofsupervisors-下属违邂上Ret三反了主管的职责，除非表明已没充分尽责:/可以将工作指派其他人负贵，但最终后果自负：-在接受短导岗位之前.必须磷保公司有充分的合爆程序.如果不合!一定要提出改进措施.公可改正好接受岗位:/如果发现违规.必率立刻行动起来沏底调直，限制当事人工作或者加强监管.11-106aw*tnGuidanceforStandardsV5.InvestmentDiligenceandreasonablebasis-耳分析报告联君做投资推荐.必须分析宏费姓济.行业、公司基本面等全部因素：/保第三方信息勤勉尽贵(四个方面
14、)：-选择外部*何要勤勉尽费(四个方面)：/如果是使用量化模型进行推介.必级MH.包括入变量.假设前提、局限性好：/崎发量化模M.需要了解偿型的方方面面，需要对模型进行焉试：集体报告如果不同意结论.但过程产i三仍然可以胃名：/HotiSSg没有尽IhCommunicationwithclients/区分事实与观点：/投资流程中的大变更要及时告知投资者.如模型.投资决策流程、投资范国.投贵限制.投费策略.关0人员改变等I/推吞可以是capsuleform,但只要投负者要.必须给出详细版*;-主要风险与局限性:Recordretention/城质版电子版保存皆口I:/IW公司的record未经批准
15、不籍带走.如果没有带走支料数据.在新公司不能发布旧公司业绩或研报，除非通过公开信息重建:/遵守当地法律定，当地没有规定的，俗金建议保存7年.M业色斯【BQ、，GuidanceforStandardsVIDisclosureofconflicts/利益冲突.指潜在可能伤害客户哦者投资公众的.主要不叶对雇主1/必须事先平实的遏育告知8/个人交易持仓：投资标的公司担任童事：与投资标的公司有业务关系(做市商、企业融费等):分析川t与投行部之间，市场部与分析棘：重大个人关系：/如果奖金激励与客户利益有冲突必绩披露：-Priorityoftransaction/Clientemployerindividu
16、al(beneficialowner).间RI时6.Conflictsofinterest;间不能太短.十天半月才合适:/区分famyaccounttjbeneficialOWner的区别；/个人交易要申报获得批准才可吸进行：Referralfees/事先披露.方使客户判断介绍是否客观.以及服务的真实成本*/最少每季度向雇主披露介绍费的性质及金H.13-106MIulaiKlGuidanceforStandardsV11Conductasmembersandcandidates，遵守考试纪律:，不得泄富考试内容和范围.但可以发表观点：/不得利用与孙会的联系为个人或公司牟取私利IReferenc
17、etoCFAinstitute,designation/正确使用CFA标志：/不可以表示因为CFA而比其他人更优秀，更胜任工作.投资业绩更好，7.Responsibilityasmembers14-106K出却曰QuantitativeMethodsFrameworkCalculateandinterpretSEEandR-squaredCalculatethepredictedVaIUeforthedependentvariableDescribeIimitatioinsofregressionanalysis161(Framework202ICF/ERM持续更新+微信：xbajun888s
18、Dummyvariables17-106FrameworkSS2QuantitativeMethodsR6Time-seriesAnalysisTrendModelsQineartrendandlog-lineartrend)AutoregressiveMOdelS(AR)+Calculateforecasts+Noautocorrelation+RegressionWithmorethanonetimeseriesFrameworkASS3QuantitativeMethodsR7MachineLearning/OverviewofMachineLearning/SupervisedMach
19、ineLearningPenalizedRegressionSupportVectorMachineK-nearestNeighborClassificationandRegressionTreeRandomForest/UnsupervisedMachineLearningPrincipalComponentsAnalysisClustering/Neuralnetworks,deeplearningnets,andreinforcementlearning)19-106*FrameworkSS3QuantitativeMethodsR8BigDataProjects/BigDataIntr
20、oduction/StructuredDataAnalysisCceptualizationofthemodelingtaskDatacollectionDatapreparationandwranglingDataexplorationModeltraining/UnstructuredDataAnalysisTextproblemformulationData(text)curationTextpreparationandwranglingTextexploration20-106Modeltraining三WQiff-IBfl*FrameworkSS3QuantitativeMethod
21、sR9ExcerptIromwProbabiIisticApproaches:ScenarioAnaIysisfDecisionTrees4andSimulations/Simulation/ComparingteApproachesMW舞！*ReviewandExploringAReviewofLevelItopks1.HypothesistestingQ/Pvalue2.Type1error/TypeII(?rrorExploringLevel11topics1.TypesofMachineLearning2.MLchallenge-Overfitting3.SupervisedML:K-
22、NearestNeighbor4.BigData-StructuredDataAnalysis5.BigData-UnstructuredDataAnalysis号曲囹新tHypothesisTestingabouttheRegressionCoefficientRegressioncoefficientconfidenceintervalB1(tfsiJClnotincludethehypothesizedvalue,rejectASignificancetestforaregressioncoefficientH0:b1=0.leststatistics:,=一hypOtheSEedVam
23、e办力=”_?Decisionrule:rejectH0if+tcriticaltortp-value:thesmallestsignificancelevelforwhichthenullhypothesiscanberejectedRejectH0ifp-value23106aSffimuHypothesisTestingTypeIerrorandType11errorTypeIerror:拒真.rejectthenullhypothesiswhenitsactuallytrue/Significancelevel():theprobabilityofmakingaTypeIerror/S
24、ignificancelevel=PfTypeIerror)TypeUerror:取伪，failtorejectthenullhypothesiswhenitsactuallyfalse/Powerofatest:theprobabilityofcorrectlyrejectingthenullhypothesiswhenitisfalse/Powerofatest=l-P(Type11error)M业SMrrlBIlHypothesisTestingH0isactuallytrueH0isactuallyfalseDonotrejectH0CorrectTypeHerrorRejectHP(
25、TypeIerror)thesignificancelevelCorreetPoweroftest=1-P(TypeIIerror)Withotherconditionsunchanged,eithererrorprobabilityarisesatthecostoftheothererrorprobabilitydecreasing.Howtoreducebotherrors?IncreasetheSampleSize.25-106，1.TypesofMachineLearningMachinelearningisbroadlydividedintothreedistinctclasseso
26、ftechniques:supervisedlearning,unsupervisedlearning,anddeeplearning.SupervisedlearninguseslabeledtrainingdatatoguidetheMLprogramtowardsuperiorforecastingaccuracy./LabeleddatasetonethatcontainsmatchedsetsofobservedinputsandtheassociatedoutputInunsupervisedlearning,theMLprogramisnotgivenlabeledtrainin
27、gdata;instead,inputs(i.evfeatures)areprovidedwithoutanycondusisaboutthoseinputs.26-106/Thealgorithmseekstodiscoverstructurewithinthedatathemselves.1W-Ilfl2.MLchallenge-OverfittingOverfittingisanissuewithsupervisedMLthatresultswhenalargenumberoffeaturesareincludedinthedatasample,resultingthatthefitte
28、dalgorithmdoesfitWRlltoHaininQdatabutDOtQRneIaliZRWRllIonewdm.ItresultsininaccuracyforecastsonoutofSamPledata1randomnessismisperceivedtobeapattern/Whenamodelgeneralizeswell4itmeansthatthemodelretainsitsRXDlanatoryDOWerwhenitisappliedtonew(Le.,t-of-sample)data.3.SupervisedMLK-NearestNeighborAK-neares
29、tneighbor(KNN).Moreccxnmonlyusedinclassification(butsometimesinregression),thistechniqueisusedtoclassifyanewobservationbyfindingsimilarities(*nearness,f)betweenthisnewobservationandtetrainingsample. KNN VWHt Ne Ohemtionf KalB. KNN WiUt New Obsmaiiour Kb5X28106w - - tra K-Nearest Neighbor Two vital c
30、oncerns The researcher specifies the value of k the hyper parameter; triggering the algorithm to lk for the k observations in te sample that are closest to the new observation that is being classified. If k is t small wil result in a high error rate, if it is too large, it will dilute the result by
31、averaging across too many outcomes./ H k is even, there Ciay be ties, with no clear winner. Analysts need to have d clear understanding of the data and underlying business to define ,similar (Or near).29-106uThepossibleerrorsinarawdatasetincludethefollowing:Incompletenesserroriswherethedataarenotpre
32、sentresultinginEiSEingdata./Themostcommonimputationsaremean,median,ormodeofthevariableorsimplyassumingzero.Invalidityerroriswherethedataare(XJtSideOfameaningfulrange.resultingininvaliddata./Thiscanbecorrectedbyverifyingotheradministrativedatarecords.InaccuracyerroriswherethedataarenotJmRasurcOftruev
33、alue.32-106Thiscanberectifiedwiththehelpofbusinessrecordsandadministrators.1U!-QllffItflStructuredDataPreparation(Cleansing)Thepossibleerrorsinarawdatasetincludethefollowing:Inconsistencyerroriswherethedataconflictwiththecorrespondingdatapointsorreality./Thiscontradictionshouldbeeliminatedbyclarifyi
34、ngwrthathersource.Non-uniformityerroriswherethedataarenotPleSentinUnidenticalformat/Thiscanberesolvedbyconvertingthedatapointsintoapreferablestandardformat.DuplicationerroriswheredDicateObSeC加ionsareDreSenl/Thiscanbecorrectedbyremovingtheduplicateentries.DataPreparation(Cleansing)Invalictity errorIn
35、consistency errorDnfLr of Rirth1 5/1 力 975切 1 M 942SalaryInCC me方bStateCredi t Card,V41 9RD$60 50 0ndp Cteness er rDp,Ton error34-i062021GEA&ERM持续更新A new variable can be created from the current variableFiltration: The data rows that are not needed for the projectm ustbeidentifiedandfiltefed357DataW
36、ranglinlJ:TransformationADatabeforetransfermaltionDataWrangling:ScalingScalingisaprocessOfadjustingtherangeofafeaturebyshiftingandchangingthescaleofdata.Herearetwoofthemostcommonwaysofscaling:Normalizationistheprocessofrescalingnumericvariablesintherangeof0.1,i(normAlixed)二3AnliHC一in/sensitivetooutl
37、iers,sotreatmentofoutliersisnecessarybeforenormalizationisperformed./ used when the distribution of the data is not known.lesssensitivetooutliersasitdependsonthemeanandstandarddeviationofthedata.37-106/Thedatamustbenormallydistributedtousestandardization.aw-aw!B5.UnstructuredDataAnalysisUnstructured
38、,texted-baseddataismoresuitableforhumanuse.Thefivestepsinvolvedneedtobemodified(thefirstfour)inordertoanalyzeunstructured,text-baseddata:1.Textproblemformulation.Theanalystwilldeterminetheproblemandidentifytheexactinputsandoutputofthemodel.2.Datacollection(curation).Thisisdeterminingthesrcesofdatato
39、beused(e.g.,webscouring,specificsocialmediasites).3.Textpreparationandwrangling.Thisrequirespreprocessingtbestream(s)ofunstructureddatatomakeitusablebytraditionalstructuredmodelingmethods./UnStRKtWeddatacanbeintheformolteximages1videos,andaudiofiles.4.TextexplorationThisinvolvestestvisualizationaswe
40、llastextfeatureselectionandengineering.5.Modeltraining38-106aw-st-TextPreparation(Cleaning)Unstructureddatacanbeintheformoftext,images,videos,andaudiofiles.ForanalysisandusetotraintheMLmodel,theunstructureddatamustbetransformedintostructureddata.ForExampleRoht Ajv 5SampleTextfromRobotsAreUSHomePageU
41、VeryhomeandbuinegSbaUldhavearbM.RawTextfromtheSourceRobotsAreUaEveryhucneandbuinesshoudlhavearobotUnstructuredTextPreparation(Cleaning)ATextcleansinginvolvesthefollowingsteps:1.RemoveHTMLtags.TextcollectedfromwebpageshasembeddedHTMLtags,whichmayneedtoberemovedbeforeprocessing.2.Removepunctuations.Textanalysisusuallydoesnotneedpunctuations,sotheseneedtoberemovedaswell.Somepunctuations(e.g.,%,$)maybeneededforanalysis,andifso,theyarereplacedwithannotations(i,
