portfolio123 virtual strategy design class by marc … virtual strategy design class by marc...

14
Portfolio123 Virtual Strategy Design Class By Marc Gerstein Topic 4D – Using Quality Factors in Your Strategies This is the final aspect of Quality, the one I think quite a few are waiting for: Earnings Quality. And that will be the final Topic in fundamentals since we’ll move next into Sentiment and then Momentum (I haven’t yet decided which sequence). But although Topic 4D may not be focusing on what many may see as the good stuff, please give it serious consideration. Something as simple as ROE can go a long way toward creating a strong tailwind that can support anything else you’re doing. And considering how much harder it is to perform out of sample than in sample, we owe it to ourselves to understand whatever tailwinds we can find. This is one of those topics that provides incredible, but often unappreciated, opportunities for creativity. In a purely Ivory Tower sense, companies with the highest quality metrics would always be the ones whose shares you’d want to own. How, after all, could one justify owning shares of other than the best firms? In the real world, though, these factors can often produce varied results. We’re always looking to the future, and when it comes to factors that can help us develop reasonable expectations of what’s to come, there often are other factors that do so with greater sense of immediacy. The key to successful use of Quality factors is to remember why we are using them and how they can work to enhance our probabilities, Getting Started Let’s start in our usual place, the DDM (Dividend Discount Model) which we re-jigger to define ideal PE as: 1 / (R – G): Lower interest rates drive R down and PE up. That’s pretty powerful. But aside from market timing (if you can do it), there’s no way for us to work with that since interest rates impact the market as a whole, and only a smaller number of stocks on an individual basis. As G (growth) rises, we know that’s good. But we’re talking about higher PEs and stronger stocks. But it’s future growth we care about. However you feel about R and G influences, they both have the virtue of being pretty direct. Quality enters the equation through often-unnoticed back doors. On the one hand, it’s part of R based on its role in the risk-premium; the higher the quality, the lower the risk so all else being equal, the lower the R. That’s an important relationship. But it does not lend itself to a 30-second CNBC sound bite, especially since

Upload: truongtuong

Post on 29-May-2018

214 views

Category:

Documents


0 download

TRANSCRIPT

Portfolio123 Virtual Strategy Design Class By Marc Gerstein

Topic4D–UsingQualityFactorsinYourStrategies

ThisisthefinalaspectofQuality,theoneIthinkquiteafewarewaitingfor:EarningsQuality.AndthatwillbethefinalTopicinfundamentalssincewe’llmovenextintoSentimentandthenMomentum(Ihaven’tyetdecidedwhichsequence).ButalthoughTopic4Dmaynotbefocusingonwhatmanymayseeasthegoodstuff,pleasegiveitseriousconsideration.SomethingassimpleasROEcangoalongwaytowardcreatingastrongtailwindthatcansupportanythingelseyou’redoing.Andconsideringhowmuchharderitistoperformoutofsamplethaninsample,weoweittoourselvestounderstandwhatevertailwindswecanfind.Thisisoneofthosetopicsthatprovidesincredible,butoftenunappreciated,opportunitiesforcreativity.InapurelyIvoryTowersense,companieswiththehighestqualitymetricswouldalwaysbetheoneswhosesharesyou’dwanttoown.How,afterall,couldonejustifyowningsharesofotherthanthebestfirms?Intherealworld,though,thesefactorscanoftenproducevariedresults.We’realwayslookingtothefuture,andwhenitcomestofactorsthatcanhelpusdevelopreasonableexpectationsofwhat’stocome,thereoftenareotherfactorsthatdosowithgreatersenseofimmediacy.ThekeytosuccessfuluseofQualityfactorsistorememberwhyweareusingthemandhowtheycanworktoenhanceourprobabilities,GettingStartedLet’sstartinourusualplace,theDDM(DividendDiscountModel)whichwere-jiggertodefineidealPEas:1/(R–G):

• LowerinterestratesdriveRdownandPEup.That’sprettypowerful.Butasidefrommarkettiming(ifyoucandoit),there’snowayforustoworkwiththatsinceinterestratesimpactthemarketasawhole,andonlyasmallernumberofstocksonanindividualbasis.

• AsG(growth)rises,weknowthat’sgood.Butwe’retalkingabouthigherPEsandstrongerstocks.Butit’sfuturegrowthwecareabout.

HoweveryoufeelaboutRandGinfluences,theybothhavethevirtueofbeingprettydirect.Qualityenterstheequationthroughoften-unnoticedbackdoors.

• Ontheonehand,it’spartofRbasedonitsroleintherisk-premium;thehigherthequality,thelowertherisksoallelsebeingequal,thelowertheR.That’sanimportantrelationship.Butitdoesnotlenditselftoa30-secondCNBCsoundbite,especiallysince

manysegmentproducersarelikelycluelessabouttherelationshipbetweenqualityandRandidealPE(assumingtheyevenknowwhatPEis).

• Also,becauseQualityimpactsconsistency(inagoodway),higherlevelsofQualitymakeitmorelikelythatresultswillbemoreconsistent,meaningthatwhateverassumptionsaboutfutureGyoudevelopthroughtheonlythingavailable(thepast),theyaremorelikelytobemoreeffectiveasabasisfordevelopingfutureexpectations.FavorableexpectationsofGarefinebuttheyareworthnothingtousiftheydon’tpanout.Again,however,thisisnotthesortofconversationyou’relikelytoseeinthefinancialmedia.

• Finally,ofcourse,Qualityisindicativeofacompany’sabilitytogenerateGinthefirstplace.

ManagingExpectationsInallthreecases,weseethattheworkingsofQualitytendtobehiddenfromthefor-dummieslevelvantagepoints.Wecouldtrytostudydataandstudysomemoreandstillmoretofindideasthatactuallywork.Orwecouldtakethepathoflessresistance:Usequalityasthemarkethandsittous,asalessconspicuousitemthatcangiveourmainstrategy,sotospeak,anextrabitofoomph.Sometimes,Qualityhelpsusmakeagoodstrategybetter.Othertimes,ithelpsusconvertapotentiallylacklusterstrategyintosomethingbetter.Anddon’tunderestimatethepossibilitythatQualitymaysimplyhelpusnarrowagood350-500stockstrategyintoanumberthat’smoreinvestable,say10-25.Thisiswhy,speakingformyselfandfrommyownexperience,I’veoftenfoundQualityfactorstoworkbestnotnecessarilywhentheyarefront-and-centerinastrategybutwhentheytakeonasupportingrole,aswasdemonstratedbackwhenwecoveredValue.Butfornow,we’llthinkofQualityasaprimarygoal.GiventherelationshipbetweenQualityandRriskcomponentofR,don’tbeatallsurprisedifyoudiscoverthataddinggood-quality-orientedfactorsintoyourmodelsreducessimulatedreturn.Itwon’thappenallthetime,butgiventhenaturalrelationshipbetweenlowerriskandlesserreturn,don’tbesurprisedifyoubumpintoit,often.Test-DrivingReturnonWhateverLets’startwithTable1,whichsummarizesacollectionofscreenbacktests.EachscreenstartedwithaPRussell3000universe,aMAXtestperiod,a4-weekrebalancingassumption,andworkedwithallstocksthatpassedthescreen(asettingof0forMax.No.PassingStocks).Also,eachscreencontainedonerule,assetforthinthetable:

RuleBasicBacktest RollingBacktest–Avg.ExcessReturn

Annl%Ret AnnlStDev Beta AllMkts UpMkts DnMktsFRank("ROE%TTM")>80 10.40 18.24 1.08 0.46 0.62 0.19FRank("ROE%TTM")<20 -0.91 34.68 1.80 -0.05 2.14 -3.46FRank("ROI%TTM")>80 10.82 18.70 1.10 0.49 0.73 0.12FRank("ROI%TTM")<20 -1.52 35.44 1.82 -0.07 2.16 -3.56FRank("ROA%TTM")>80 10.49 18.88 1.10 0.47 0.71 0.08

FRank("ROA%TTM")<20 -1.59 35.68 1.84 -0.07 2.22 -3.65FRank("ROE%5YAvg")>80 10.07 17.66 1.04 0.42 0.44 0.38FRank("ROE%5YAvg")<30* 3.91 25.88 1.36 0.15 1.15 -1.41FRank("ROI%5YAvg")>80 10.49 18.72 1.10 0.45 0.64 0.16FRank("ROI%5YAvg")<30* 1.75 30.06 1.63 0.03 1.71 -2.60FRank("ROA%5YAvg")>80 10.59 18.77 1.10 0.45 0.63 0.18FRank("ROA%5YAvg")<30* 1.40 30.05 1.63 0.01 1.74 -2.69*Thresholdsetat30toproduceareasonablenumberofstockspassingthescreenThetableshowsussomeinterestingthings.

• WhileROE,ROIandROAarecomputeddifferentlyandprovidedifferentinformation,wecanassume,unlessotherwisedictatedbytheuniqueneedsofaparticularstrategy,thattheinvestmentimplicationsofthethreeratiosis,forallpracticalpurposes,thesame.(AndwecanpresumelikewiseregardingthecountlessvariationsthatcanbefoundonInvestopedia,Wikipediaandwho-knows-how-manyothersources.)Weshouldnotbesurprised.Ifwelookattheformulas,weshouldexpectahighcorrelationinrankingsfromoneratiotoanother.

• Inbacktest,thereislittledifferencebetweenTTMand5-yearresultsandthisistobeexpectedgiventheoverallbig-picturepersistenceofthesereturnitems.Butthatdoesn’tmeanwecanflipacoinwhenmodelingforward,aswemustdowhenwethinkaboutrealmoney.Onewhoownsonly10-20stocksneedstobesensitivetotheexceptionsthatgetpaperedoverinlargerstudiessuchasthis.Soregardlessofwhatacademic-typetestsshow,thereshouldstillbeagoodreasonforpickingTTMorafive-yearaverage(thereisneveranacceptablereasonforeventestingmuchlessusingaQnumberinthecontextofQuality;ifanything,itwouldbeaMomentumfactoranditwillbediscussedfurtherwhenwereachthattopic).Generally,amorereturn-orientedmodelcanleantowardthemorehere-and-nowTTMfactorwhileonewhoismoreinterestedinrisk-controlcouldworkwithlonger-termaverages.

• Thegapsbetweenbestandworstraisetheprospectthatwecanaccomplishmuchinourworkevenifwedonothingelsewiththeseitemsotherthantoscreenouttheworst.Thisisoneofcountlessreasonswhyitsvitalthatyounotobsessoverrankingsystems.It’samazinghowmuchyoucanaccomplishevenwithplain-vanillasystemsifyoucanrunthemagainstpre-qualifiedsub-universesthathavealreadyidentifiedandweededoutpotentialtrouble.Onewaytobeatthemarketistoidentifyandoverweightwinners.Anotherequallyeffectivewayistoidentifyandunderweightthedregs.Thatsaid....

• Don’tbeafraidoflowqualityifyouunderstandwhatitmeansandwanttoworkinthatmanner.Notice,thedifferencesinrollingup-anddown-markettestsforthejunkgroups.Thereisalotofdownsiderisk;that’sobvious.Butthereisalsoseriousupsidepotential.Thishappensbecauselowqualityisassociatedwithpoorconsistencyandifyouarebeingaggressive,that’swhatyouwant–poorconsistency.Yourjob,insuchacase,istousescreeningrulestotrytolimityoursub-universetosituationsmorelikely

thannottocapturethegoodpartofinconsistency.Technicalfactors,momentumandsentimentcanhelpalotinthisarea.

• Butthroughitall,ifyoujustwantareasonablypositiveresultsandwithnomorethanreasonablelevelsofrisk,useofROE,ROAorROIcanputaheavytailwindatyourback.This,bytheway,isabigexplanationforhowWarrenBuffettgottobeWarrenBuffett.Hemightnotbeabletocreatea90%-alphap123sim.Butstill,hedidprettywellforhimself.

FromtheBigPicturetoInvestabilityLikeacademicstudies,theoneaboveworkedwithlargeswathsofalargeuniversethatidentifiesaggregatecharacteristics.Westillhavetoworkourwaydowntomanageable-sizeportfolios.Sothenextsetofexperimentswillexaminewhatwecanexpectreturnstoaccomplishforusifwelimitourpositionsto15.InallcasesI’mgoingtoworkwithROE%TTM.Ifyouwanttorepeattheexperimentswith5Yearand/orwithROIorROA,goforit.I’mgoingtostartwithaverysimplescreenagainstthePRussell3000universe(It’sbeenawhilesinceIsaidthissoarefreshercouldn’thurt:PleasedesignyourstrategieswithauniversenobroaderthanthePRussell3000.Yourgoalistocomeupwithastrategythatcanworkwithrealmoney,nottoproduceeye-catchingsims.Soyoudonothelpyourselfifyoudesignusingamarshmallowuniverse.IfyougetsomethingthatworkswiththePRussell3000,youcanalwaysgobacklatertoswapintheAllFundamentalsuniverseandpopinsomeliquidityrules.)ScreeningRule:FRank("ROE%TTM")>80QuickRank:ROE%TTMhigherisbetter,pickTop15Here’stheresult:Figure1

RollingTestAvg.Excess4-weekReturn:0.27%inallperiods,0.00inupperiods,and0.69indownperiodsTheresultispositive,butaheckofalotlessappealingthanwhatwesawinouracademic-styletesting.ThebasicnumbersactuallyseemOK,butthepicturealonesuggestsalotlessappeal.Andtherollingnumbersmakeitclearthatwe’rereallydependentonbadmarkets,whenthismodelmighthavesomedefensiveappeal.Butoverthelongtime,wehavemoregoodperiodsthanbad.SoitseemsthatdespitetheaggregatevirtuesofROE,whenitcomestomanageablesizeportfolios,weneedtodomorethanrankandcountdownfromthetop.Whatwillfollownowisasetofiterations,butthesearen’titerationsinthesensediscussedbystatisticians.We’renotgoingtojustchangethingsarounduntilwefindaresultwelikeandthensay“We’redone.”Eachiterationwillbemotivatedbyaspecificallystatedgoalbasedonfinancialtheory,specificallyasearchforwhatsortofdatapointislikelytopointusinthedirectionofhighROEsthataren’tflukybutreal.Ifrationallyjustified,youcanhaveasmanyiterationsasyouwant,limitedonlybyyourimagination.I’llofferseveniterationshereinordertoillustrateathought-and-feedbackprocess.Onceyouseewhat’sbeingdone,youshouldbeabletocontinueonyourown.Let’sthinkoftheaboveasIteration#1andmoveon.Iteration#2IknowROE%TTMworksintheaggregatebecausefinancialtheorytellsmeitworks,andithelpsthatIsawitinthestudywhoseresultswereshowninTable1.Buttomakeitworkinthecontextofaninvestableportfolio,Ineedtodomorethanpickfromthetop.Knowing,asIdo,thatROEworksbecauseofthewayitimpactsGandRintheDDM,IreasonthatROEcan’t

accomplishwhatIhopeitwillifit’sintheprocessoftrendingdownward.SoI’llexperimentwiththis:ScreeningRules:FRank("ROE%TTM")>80 ROE%TTM>ROE%5YAvgQuickRank:ROE%TTMhigherisbetter,pickTop15Here’stheresult:Figure2

RollingTestAvg.Excess4-weekReturn:0.62%inallperiods,0.99%inupperiods,and0.05%indownperiodsThestandardbacktestisfine.Iexpectedimprovement(throughtheadditionofasecondrulethattriedtoweedoutsituationswhereROE,althoughgood,istrendinglower)andgotit.Butitisateenybitimperfect–lacklusterperformanceintherollingdownperiods(mostlikelynotdifferentfromzerotoasignificantdegree).It’snottheendoftheworld;wedohavemoreupthandownperiods.Butalthoughwecanlivewithwhatwehave,whystopsoquickly.Thegoaliftestingistolearnwhatwecanandcan’tdoaswetranslateideasintop123lingo.Wecanalwayscomebacktothisandsettleinifthat’swhatweultimatelydecidetodo.Iteration#3ScreeningRules:FRank("ROE%TTM")>80 OpMgn%TTM>OpMgn%5YAvgQuickRank:ROE%TTMhigherisbetter,pickTop15

OnewaytoaddresstrendsinROEistodrilldowntoitscomponentparts,oneofwhichismargin.AsweknowNetMarginistheversionwe’dconsiderifwewanttostrictlyreplicatetheDuPontframework.Butforourforward-lookingpurposes,wecanswapinanyothermarginthatwethinkwillbetterilluminatethecompany’spotentialfuture.Let’stryOperatingMargin.It’shighenoughintheincomestatementtoeliminateissuesinvolvingcapitalization,specialitems,etc.butunlikegrossmargin,we’resparedtheburdenofdecidingdifferencesinhowcompaniesmightallocateexpensesasSG&AorCOGS.AswithIteration#2,weseekaTTMfigureinexcessofthefive-yearaverage.Figure3showstheresults.Figure3

RollingTestAvg.Excess4-weekReturn:0.49%inallperiods,0.99%inupperiods,and-0.30%indownperiodsIt’samildstepbackwards.Ifwereallywanttoworkwithmargin,we’dneedtoconsiderwhythismovedusinthewrongdirection.Maybeoperatingmarginwasn’tthebestchoice.MaybeTTM>5Yistoolazyawaytoarticulateatrend;perhapsweneedtogetmoregranularandworkyearbyyear.(WemightthinkofthiswithrespecttoIteration#2aswell.)Maybeweshouldaddsomethingrelatingtoturnover;eitherwedrilldownintoDuPontcomponentsforreal,orwedon’t.I’llleaveittoyoutoworkfurtheralongtheselinesifyouwish.What’simportanttonote,here,ishowyoumovefromoneiterationtothenext.Itisn’tamatterofplugginginonethingoranothertocoverallpossiblebases.Youhavetotalktoyourself,askwhythelastthingwasn’tasgoodasyouexpectedandwhatitmighttaketoaddresstheshortcoming.Iteration#4

ScreeningRules:FRank("ROE%TTM")>80 FRank("DbtTot2CapTTM",#industry)<50QuickRank:ROE%TTMhigherisbetter,pickTop15Weknow,fromtheDuPontframework,thatleverage/debtispartofthepackage.Wealsoknowthatallelsebeingequal,lessdebtisbetter(becarefulthough,allelseisaverybroadconcept).Wealsoknowthatevenifweholdoperatingincomeconstant,wecanboostROEbymakingdebtabiggerpercentofthecapitalstructure.Soperhapswecanmitigatethepotentialbalance-sheetriskposedbyahigh-ROEmodelifweincorporatealeverage-relatedfactor.Sohowfarshouldwegoinlimitingleverage.WehavetostartsomewheresoI’llpickanFRank<50approach.Andthiswouldbeaterrifictimetospotlightanimportantissuethatwehaven’tyetaddressedbutwhichisalwaysonthetable:Shouldwedoacompletesort,orshouldwesospecializedsortsbasedonindustries,sectors,etc.Thereisnoinherentlyrightorwronganswer.Ifwewanttruly“better”companies,weshoulduseanindustry-typesort.Beingabletooperatewithlessleveragecomparedtoothersinthesamebusinesswithsimilarbalancesheetneedstellsussomething.Ifhavingasector/industrydiversifiedportfolioisimportanttous,weshouldworkwithsuchsorts.Ifyou’reaprorunninglargeaccounts,youprobablydoneedtogothisway.Ifyou’reanindividual,however,youcanaffordtorefrainfromthissortofthing.Sectorimbalanceisfine,ifyourmodeltiltsyoutowardbettersectors.Abasicsortagainsttheentiregroupwillworkforyouhere.(Diversificationcanbeaccomplishedthroughmultiplepositionsanduseofmultiplefactors.Sectordiversificationisbasedheavilyonstereotypeandcanincreaseriskifitforcesyoutoincreaseexposuretohigh-riskbusinesses.)Figure4

RollingTestAvg.Excess4-weekReturn:0.49%inallperiods,0.86%inupperiods,and-0.11%indownperiodsThat’snotsohot.It’snothorrible,butwe’veseenbetter.Icheatedandtookoutthe#industryparameter,ranthetestagain,andgotresultsclosertowhatwesawwithIteration#3,whenwefocusedonoperatingmargin.Wedefinitelypaidapriceforhavingtriedtobegoodcitizenswhenitcomestoindustryexposure.Butaswithoperatingmargin,wefindthatifwewanttosupportabasicquestforhighROEwithexaminationofindividualDuPontcomponentswe’llhavetoworkharderatchoosingandarticulatingthemthanI’vesofardone.Iteration#5ScreeningRules:FRank("ROE%TTM")>80 Rating("Basic:Quality")>90QuickRank:ROE%TTMhigherisbetter,pickTop15Timeforachangeofpace:Let’snottrytosupportthebasichighROErulewithpiecemealDuPontconsiderations.Instead,let’sgowholehogwithalotofthematonce.Henceuseofthe“Basic:Quality”rankingsystemaspartofascreeningrule.Thefactorsarevisibletoyousoifyoucheck,you’llseeitcoversalotofterritory.Figure5

RollingTestAvg.Excess4-weekReturn:0.42%inallperiods,0.56%inupperiods,and0.19%indownperiodsWe’vemadeprogress,verysmallprogressbutprogressnonetheless.That’sseenintherollingtests.Thisraisestheprospectthatwecoulddomorealongtheselines,perhapsbytappingintomorefactors,ifnotthroughuseofrankingsystemsthanbyuseofmorescreeningrules.Also,Icouldrelaxthe90threshold.Remember,wecanmakeprogresssimplybyeliminatingdogs.Still,wearegettingabitQualityobsessedhere....Iteration#6ScreeningRules:FRank("ROE%TTM")>80 Rating("Basic:Value")>90QuickRank:ROE%TTMhigherisbetter,pickTop15Here’sachange-up.We’llleavetheoriginalROEfactorsastheyare,acceptthemforwhattheyareproandcon,andbroadenourstrategytogoforQualityatareasonableprice,theGreenbattphilosophy.Greenblattdiditoneway.Thisiterationillustratesanother.Figure6

RollingTestAvg.Excess4-weekReturn:0.96%inallperiods,1.94%inupperiods,and-0.56%indownperiodsOurinterestispiqued.Returnisup.We’renotyethome;volatilityistroublesome.AndthatisconsistentwiththeDDMscript(loweridealvaluationratiosareassociatedwithgreaterrisk),andwithcommonsense(manylowratiosaresuchbecausetheydeservetobelow;i.e.becausethecompaniesarebad).ButValueisoneofthemostwellestablishedreveredapproaches.Soratherthanjustrunawayfromtherisk,let’sseeifwecantameit.Iteration#7ScreeningRules:FRank("ROE%TTM")>80 Rating("Basic:Quality")>90

Rating("Basic:Value")>90QuickRank:ROE%TTMhigherisbetter,pickTop15Wedouble-downonouruseofRatings-basedBuyrules;wewantstockshighlyratedforValueandwecontrolriskbyinsistingthattheybestronginQualitytoo,abroadrangeofQualityandnotjustROETTM.Figure7

RollingTestAvg.Excess4-weekReturn:1.98%inallperiods,3.19%inupperiods,and0.09%indownperiodsWellthat’seye-catching.Wecertainlygotahigherreturn.Butohthatvolatility!Whattheheckishappening?Theanswerisprettyeasytosee.We’reover-screening.Themostrecentrunofthescreenproducedonly7passingstocks.Andweweresimilarlylow,andlower,onmanyotheroccasions.We’regettingtoocrazyandoverlyexposingourselvestotoomanythingsthathavethefunctionalequivalenceofrandomness.That’seasytofix...Iteration#8ScreeningRules:FRank("ROE%TTM")>80 Rating("Basic:Quality")>75

Rating("Basic:Value")>75QuickRank:ROE%TTMhigherisbetter,pickTop15Wejustlowerthescreening-ruleRatingthresholdsfrom90to75(thelatternumberisn’tsacrosanct;Ipulleditoutofthewind).Wenowhave85stocksthatpassthescreen.Maybethat’sfine(weultimatelysortourwaydownto15)ormaybeweneedtolowerthethresholdabitmore.Wecanworkwithit.Butlet’sseeifwe’reatleastontherighttrack.

Figure8

RollingTestAvg.Excess4-weekReturn:0.98%inallperiods,1.45%inupperiods,and0.25%indownperiodsBingo.Thetrainismovingforwardagain.Wedoseeprolongedperiodsofsidewaysmovementandthensomeseriousrallies.That’sOK.Patienceisn’ttheworstthingintheworldiftheideamakessense.Andthefactthatwe’rebringinglowvaluationratiosintothepicturesuggestsitsunavoidable(valueisaninformationarbitragestrategy,theideabeingtolookforsituationsinwhichthemarkethasmisjudgedRandG,andsometimesittakesawhileforthattoplayout).WestillhaveIteration#2inourpockets,sowedon’thavetoforceourselvestosettleforthis.Wecan(theoverallreturnsarefine)butwemaynothaveasmuchpatienceaswewishwedid.It’sstillanopenquestion.Morecanbedone.Buthopefully,younowseehowyoucanpursueanswers.It’sAboutTheProcessOneobviousgoalofthisTopicwastogetyouthinkingthekindsofthoughtsyouneedtobethinkingasyouworkwithQuality.Beyondthat,though,Iwantedtoillustratemorecloselytheprocessofdevelopingandtestingastrategy.Thedifferencebetweencurvefitting,dataminingetc.,practicesthatcanmakeforgreatsimsbutrottenreal-moneyresults,isnottobefoundinstatisticalconcepts(robustness,etc.).Actually,arobustmodelmeansyoudidabetterjobofpredictingthepast,soinsteadofhavingamess,youmayhaverelevateditintoaperfectmess.Butamessisstillmess,andThat’snotwanttobedoing.

Alegitimatelydesignedmodelisonethatspringsfromrationalideas.Ifyou’redatamining,20iterationscanbetoomany.Ifyou’redoingitright,2,000canstillleaveyouwithroomformanymore.Andyouknowyou’redoingitrightifyoucanexplaintoyourselfwhyyouarechoosingtotestsomethingandunderstandingwhythetestmayhavesucceededorfallenshort.Andbytheway,didyounoticethatnoneofthesetestswasdisastrous.Thisisimportant.Ifyourtestsarederivingfromsensibleconcepts,youshouldknow,goingin,thattheresultswon’tbehorrifying.Non-horribleoutcomesshouldbepresumedevenbeforeyouclickto“Run”atest.Whatyoulearniswhetherit’sgoodenough.Also,noticetheDDMmentions.Ididn’tspendalotoftimeonthem.Butit’simportanttorecallhowivory-towertheDDMreallyis:D/(R-G)inaworldwheredividendsmaynotbepaid,whereanon-infiniteGmightmakeforanegativeP,etc.ThisTopicillustrateshow,ifwethinkofDDMasaconceptualanchorratherthanausableformula,wecanbuildsensiblemodelsthattestdecentlyrightfromthefirstpassandwhichstandagoodchanceofsucceedingwithrealmoneyorbeingaspringboardforsomethingwecansuccessfullyuse.Andhopefully,Table1motivatesyoutoexplorethefullrangeofQualityandwhatyoucandowithit.Thenexttopic,thelastinQuality,willbeEarningsQuality.