QuantitativeInformationAnalysis_南京大學中國大學mooc課后章節(jié)答案期末考試題庫2023年Whenoptimizingamodel,weshouldalwaysconsideratrade-offbetweenmodelinterpretationandmodelaccuracy.

參考答案:

正確

Thestagesofdataanalyticsusuallyincludefeatureselection,modelingandtesting,modelevaluationand.

參考答案:

optimization

WewillusePython,Rand____inthisclass.

參考答案:

Stata

Whatistheultimatepurposeofdataanalysis?

參考答案:

Togaininsightsandhelpwithdecisionmaking

TheoutputofLDAwillbethematrixandmatrix.

參考答案:

word-versus-topics;document-versus-topics

InthealgorithmforLDA,anewtopicisassignedtoeachword,wherewechoosetopictwithprobability.

參考答案:

P(topict|documentdj)

*

P(worddj|topict)

LDAcanbeutilizedtogeneratetextvectors.

參考答案:

正確

A_________isatypeofstatisticalmodelfordiscoveringtheabstract“topics”thatoccurinacollectionofdocuments.

參考答案:

topicmodel

Whichoneofthefollowingcategoriesdoestherankingofastudent'sscore(e.g.,2)belongto?

參考答案:

Ordinal

Whichoneofthefollowingcategoriesdoestheaveragescoreofacourse(e.g.,85.2)belongto?

參考答案:

Ratio

______istheconditionthatoneisexposedtoincausalinference.

參考答案:

treat

WhatmightbethepotentialdrawbacksofRandomControlTrial(RCT)?

參考答案:

Timeconsuming_Sometimesnotrandomizable_Budgetconsuming

isthesumoftheabsolutedifferencesoftwovectors’correspondingfeatures.

參考答案:

Manhattandistance

Whichofthefollowingistextclassification?

參考答案:

Identifyingwhethertheemotionofadocumentispositiveornegative._Identifyingadvertisementsinemails.

Euclideandistanceeliminatestheeffectoftextlength.

參考答案:

錯誤

_______________looksforpreviouslyundetectedpatternsinadatasetwithnopre-existinglabelsandwithaminimumofhumansupervision.

參考答案:

Unsupervisedlearning

Whichoneofthefollowingcategoriesdoesthegenderofastudent(e.g.,female)belongto?

參考答案:

Nominal

Whichoneofthefollowingcategoriesdoestheageofastudent(e.g.,19)belongto?

參考答案:

Interval

Informationanalyticsinclude:

參考答案:

Allofabove.

focusesonthepolarityofagiventext.

參考答案:

fine-grainedsentimentanalysis

Forobjectivityanalysis,theautomaticmethodswillclassifytextinto.

參考答案:

objectiveandsubjective

Rule-basedsystemsareveryna?vesincetheydonottakeintoaccounthowwordsarecombinedinasequence,nordotheytakeintoaccountthedifferencemeaningsofthesamewords.

參考答案:

正確

Wearestudyingtheeffectofcollegeeducationonone’sfuturesalary.SupposeTomhasgonetocollegeandhissalaryis$5,000permonth.WhichofthefollowingisTom’spotentialoutcome(potentialsalary)andobservedoutcome(observedsalary)?

參考答案:

Tom’searningsifhehasnotgonetocollege;Tom’spresentearnings

Howcanquantitativedataanalysisbeappliedinhealthcare?Chooseallappropriateoptions:

參考答案:

Useneuralnetworkmodelstostudythedatafrom

wearabledevicesandpredictdisease_Usetextanalysistobuildelectronicmedicalrecord_Usenetworkanalysistostudythespreadofanepidemic_Useregressionanalysistofindtherelationshipbetweenhealthcareexpenseandage

Howcanquantitativedataanalysisbeappliedinsports?Chooseallappropriateoptions:

參考答案:

Analystshelpteamsidentifytrendsthatmayimproveon=courttacticsorpracticehabits_Athletesusewearabledevicestofindhowfatiguewouldaffecttheirperformance_Somewearabletechnologylikehelmetscouldhelpathletesminimizeinjurythroughreceivingdata_Acoachusedatatodeterminethebestplaytocallinaspecificsituation

Bysomeestimates,businessesandfinancialinstitutionsuseonly0.5%ofavailabledata.Themostprobablecausesforthisproblemshouldbe___.

參考答案:

thatthehugevolumedataandnewdataproducedatahighspeedmakedataanalyticsmoredifficult_difficultyindatapreprocessinganddataorganization

Quantitativeinformationanalysisismorereliablethanqualitativeinformationanalysis.

參考答案:

錯誤

Whichoneofthefollowingdoesnotbelongtothedescriptivestatisticsofonequantitativevariable:

參考答案:

Distribution

Fortf-idfweightedbagofwords,rarewordsthatonlyappearinafewdocumentswillhaverelevantlylargerweight.

參考答案:

正確

Whatarethecriteriathatacausalrelationshipmustfulfill?

參考答案:

Thecauseandtheeffectmusthavetemporalorder._Thecauseandeffectmustbecorrelatedwithoneanother._Theobservedempiricalcorrelationbetweenthetwovariablescannotbeexplainedawayastheresultofathirdvariable._Repeatedpracticeofthecausealwaysleadstothesameeffect.

Theexistenceofoutputdeterminesthetypeofstatisticallearning.Theyare___and___.

參考答案:

supervisedlearning,unsupervisedlearning

Whenwemakepredictions,wecommonlyuseclusteringandclassification.Forexample,ifIwanttodecidethebreedofdogsusingtheirbiologicaldata,Iwilluse______.

參考答案:

classification

TheMattheweffectisexamplifiedinthislectureas:alimitednumberofpapershavea__numberofcitations:

參考答案:

large##%_YZPRLFH_%##big##%_YZPRLFH_%##great

Whichofthefollowingcannotbeusedforbivariatedataanalysis:

參考答案:

One-dimensionaldistribution

Insituation_,wewillusePearson’scorrelationcoefficient.

參考答案:

Continuous,normallydistributed,lineardata

Intheexampleaboutconfidenceintervalinthislecture,whydowedrawsamplesfor1000times?

參考答案:

Decreasethenegativeeffectfromunusualdatapoints

Whichofthefollowingisnotamethodusedformultivariatedataanalysis:

參考答案:

Singlescatterplot

Whichofthefollowingvalueshowsapositivecorrelation:

參考答案:

1.0

PCAcouldbeusedfordatavisualizationanddimensionreduction.

參考答案:

正確

PropensityScoreisoftencalculatedwitharegression.

參考答案:

Polynomial

Thefollowingtableliststhepropensityscoreforthetreatmentandcontrolgroup.Whendoinggreedymatchingwithoutreplacement,whichsubjectinthecontrolgroupismatchedtoSandyinthetreatmentgroup?【圖片】

參考答案:

Jack

Whenstudyingtherelationshipbetweencollegeeducationandfuturesalaries,whatmightbepossiblefactorsthatcauseselectionbias?

參考答案:

IndividualIntelligenceLevel_FamilyBackground_Theeffortonecandevotetohiscareer

Anisanempiricalinvestigation,inwhichtheobjectistoelucidatecauseandeffectrelationshipsinsettingsinwhichitisnotfeasibletousecontrolledexperimentation.

參考答案:

observation

Machinelearningisthesameasstatisticallearning.

參考答案:

錯誤

Mostbasketballplayersareextremelytall.Therefore,playingbasketballmakesonegrowtaller.Whichmistakeisthestatementsubjectto?

參考答案:

ReverseCausality

Selectionbiascancauseendogeneityissue.

參考答案:

正確

Theassumesthatthetreatmentgroupwouldhavethesametimetrendasthecontrolgroupabsentfromthetreatment.

參考答案:

Paralleltrendsassumption

_____isamulti-dimensionaldatainvolvingmeasurementsovertime.

參考答案:

Paneldata

Whichofthefollowingstatementforomittingtheexplanatoryvariablesisnottrue?

參考答案:

Explanatoryvariable

x

inyourmodelisalwayscorrelatedwithestimatedresidualsifyouomittedsomeexplanatoryvariables.

Inmeasurementerror,thereexistsdeviationbetweenthesamplevalueandtherealvalue,if【圖片】istheobservedvalue,【圖片】istherealvalueand【圖片】isdeviationthenwhichofthefollowingisnottrue?

參考答案:

Covariancebetween

andresidualisnolongerzero.

Wecallthephenomenonofscientistsmovingfromadevelopingcountrytoadevelopedcountryas:

參考答案:

BrainDrain

Whatquasi-experimentmethodsareusedinthisresearch?

參考答案:

PSM-DID

Whichofthefollowingbelongstoselectionbias?

參考答案:

Aftermoving,onewillworkmoreefficientlybecausetheyareexposedtonewenvironmentandnewideas.

Inthisresearch,whichofthefollowingtwoindicatorsareusedtomeasurescientists’ability?

參考答案:

Thenumberofpaperstheypublishedperyear_Theirchancesofbeingcorrespondingauthors

Whichofthefollowingdescribeofendogeneityisnottrue?

參考答案:

Endogeneitywillnotdestroytheuniformityparameterestimates.

Whatkindofregressionmodelcanweusetoexplorerelationsbetweengenderandcitationscounts,giventhatthenumberofcitations(thedependentvariable)ishighlyskewedandthestandarddeviationofthenumberofcitationsissimilartoitsmean?

參考答案:

NegativeBinomialregressionmodel

Whichofthefollowingpublicationdependsstronglieronearlierpublications(bothreceived5citationsandhave3references).

參考答案:

A

Whichofthefollowingfigurehasabroadercitationimpact?【圖片】

參考答案:

B

Forprincipalcomponentanalysis,whichofthefollowingstatementisnottrue?

參考答案:

TheindependentvariablesafterusingPCAalwaysmakesense.

Thisisamatrixmultiplicationformultiplelinearregression,【圖片】,ifnisthenumberofindependentvariablesand【圖片】isann×1columnvector,pisthenumberofobservations,thenwhichofthefollowingstatementistrue?

參考答案:

isann×(p+1)matrix.

WhichofthefollowingmightviolatetheinternalvalidityofRDD?

參考答案:

Theobservationsonthetwosidesofthecutpointaretreateddifferentlybecauseoffactorsotherthanthetreatment.

WhenvisualizingRDD,thelargerbinwidthsweselect,thenoisewegetandtheinformationwelose.

參考答案:

less,more

InRegressionDiscontinuityDesign,therunningvariablecanmustbenumeric.

參考答案:

正確

Accordingtothisfigure,wecanarguethatcollaborationenhancesproductivity.【圖片】

參考答案:

錯誤

Inthisresearch,whichofthefollowingisusedastherunningvariableforRDD?

參考答案:

Theevaluationscorefortheapplication

Thescreeningeffectisthesingleonefactorthatcausedthegapbetweennear-winsandnear-misses.

參考答案:

錯誤

________effectsuggeststhatearly-careersuccesshelpsbringfuturevictoriesbecauseitnotonlybringscommunityrecognition,futureattentionandacademicresources,butalsobolstersself-confidence.

參考答案:

Matthew

_________mechanismsuggestthatifearly-careerfailuresscreenoutless-determinedresearchers,earlysetbacksamongthosewhoremaincould,perhapscounterintuitively,becomeamarkerforfutureachievement.

參考答案:

screening

Basedonwhatyouhavelearnedabouttypesofregression,whichofthefollowingstatementistrue?

參考答案:

PLSregressioncanbeusedforseveralquantitativedependentvariables.

Thefirstharticlesrankedbythecitationcountsindescendingorderformtheh-core,thenh-tailis

參考答案:

alltheotherarticlesoutsidetheh-core

WhichofthefollowingstatementforgoodnessoffittestforOLSisnottrue.

參考答案:

Goodnessoffittestisusedtodescribethevolatilityof

Yi

relativetoitsfittedvalue.

Theh-coreconsistspublication

參考答案:

A,B,C,D,E

Theextendedh-coreincludingpublications

參考答案:

A,B,C,D,E,F

Whichofthefollowinghasnochancestoaffectthevalueoftheh-index?

參考答案:

Increasingthenumberofco-authors

Thecentralfeatureofgrangercausalityis.

參考答案:

Predictability

Iftimeseries【圖片】canbeusedtopredict【圖片】,whichofthefollowinghypothesismuststandforthefollowingformula?【圖片】

參考答案:

Atleastoneof

isnotzero.

Grangercausalityonlyrevealscorrelation,notcausality.

參考答案:

正確

Weusuallyuse_totestPearsoncorrelationcoefficient.

參考答案:

t-test

Weusuallyuse__totestSpearmanrankordercorrelationcoefficient.

參考答案:

z-test

WhichofthefollowingstatementsofSpearmanrankordercorrelationcoefficientandKendallrankcorrelationcoefficientistrue?

參考答案:

WechangeoriginaldatatorankpairofdatainSpearmanrankordercorrelationcoefficienttoeliminatetheinfluenceofoutliers.

AccordingtotheruleofOLS,the__needstobethesmallest.

參考答案:

residualsumofsquares

Inthefollowingequation【圖片】,【圖片】is__.

參考答案:

residual

_____isthemostwidely-acceptedindextojudgetheacademicimpactofaresearchpaper.

參考答案:

Citation

Whendecidingthedirectionofcausalityusingthefollowingtwoformulas,whatresultsindicateunidirectionalgrangercausalityfromcitationstodownloads.【圖片】【圖片】

參考答案:

Atleastoneof

doesnotequal0whileall

isequalto0.

Whendecidingthedirectionofcausalityusingthetwoformulasinthelastquestion,whatresultsindicatebidirectionalcausality.

參考答案:

Atleastoneof

doesnotequal0andatleastoneof

doesnotequal0.

Withtheresult,wecandrawaconclusionthatdownloadsisagoodindicatorforfuturecitationimpact.【圖片】

參考答案:

錯誤

Whichofthefollowingstatementaboutregressionanalysisisnottrue?

參考答案:

Regressionanalysisisomnipotent.

Onepotentialflawoftheh-indexisthatpublicationsandcitationsarenot_____oneachother.

參考答案:

independent

Aninformationnetworkconsistsof_______.

參考答案:

bothAandB

Anacademiccollaborationnetworkisonetypeofsocialnetwork.Insuchanetwork,anodewithdegreetwomeansthat:

參考答案:

Ascholarhascoauthoredpublicationswithtwootherscholars.

InGraph1,wecouldfindaEulercircuit:Startingfromnodea,wecouldtravelthroughalltheedgesincircuit(a,e,c,b,e,d,a)withoutrepetition.WhatabouttheSevenBridgesofK?nigsberg?CanyoufindaEulercircuitinit?【圖片】【圖片】Graph1SevenBridgesofK?nigsberg

參考答案:

錯誤

Howmanynodesarethereinthisgraph?Howmanypathsaretherebetweennodeaandnodec?【圖片】

參考答案:

9,11

FillintheblankoftheadjacencymatrixAofthegraphbelow.【圖片】【圖片】

參考答案:

A31=1,A42=2

Thisisaweightedgraphmodelofanairlinesystem,wheretheweightonanedgeistheleastpriceavailableforaflightbetweenthetwocities.FindaroutewiththeleasttotalairfarebetweenSanFranciscoandMiami.【圖片】

參考答案:

SanFrancisco–NewYork-Miami

Acompletegraphonnnodes,denotedbyKn,isagraphthatcontainsexactlyoneedgebetweeneachpairofnodes.Thereisnoself-circleinacompletegraph.K1,K2andK3areshownbelow:【圖片】SohowmanyedgesarethereinaK4?HowaboutKn?(n>=2)?

參考答案:

6,

Theprocessofdrawingaconclusionaboutacausalconnectionbetweenthecauseandeffectiscalled______.

參考答案:

causalinference

Whichofthefollowingterminologyisthesameasreproducibility

參考答案:

Reliability

Normalizationwithz-valuecanbeusedfor

參考答案:

Normallydistributeddata

Theclosenesscentralitycanmeasurethe_______ofanetwork.

參考答案:

efficiency

Thebetweennesscentralityisthenumberoftimesanodeneedsagivennodetoreachanothernode,whichmeansitisthenumberofpathsthatpassthroughagivennode.

參考答案:

shortest

Whichofthefollowingconceptsisnotrelevanttothediversityindicators

參考答案:

Modularity

Anindicatorisamathematicalexpressionthatcanbeappliedto________data.

參考答案:

numerical

Balancedescribesthenumberofdistinctivecategories.

參考答案:

錯誤

WhichofthefollowingstatementofclassicalanalysisandEDAisnottrue:

參考答案:

InEDA,weestimatethemodeldirectlyafterpreparingdata

Doestheh-indexmeasurehowGOODascientistis?

參考答案:

錯誤

Readthefollowingtableandanswerquestions2-4Thehindexofthisauthoris.【圖片】

參考答案:

5

Sentimentanalysisisakindof.Itutilizestexttechniques.

參考答案:

Supervisedlearning;classification

Whatistheaveragepathlengthofthisgraph?(Hint:listtheshortestpathbetweeneachpairofnodesandcalculatetheaverageofthem.Refertothedefinitionofit.)Whataboutthediameter?【圖片】

參考答案:

23/15,3

Inasocialnetwork,whichofthefollowingwouldoneexpecttobetrueaboutthedegreesofitsnodes?

參考答案:

Avarietyofdegreesistobefound

Selectthefeaturesofasmallworldnetwork.

參考答案:

Ahugesize_Nohub_Sparse

Accordingtotheformulaofaverageclusteringcoeffcient,calculatetheaverageclusteringcoefficientofthisgraph.【圖片】

參考答案:

1

Ahubinanetworkisanodewithhighdegree.

參考答案:

正確

Hubsaremorelikelytobemoreimportantthanothernodes.

參考答案:

正確

Hubsonlyappearinnetworkswithmorethan2nodes.

參考答案:

正確

NetworkAhas50nodesand100links.TheaveragedegreeofnetworkAis_.

參考答案:

2

Howmanyconnectedcomponentsarethereinthegraphbelow?【圖片】

參考答案:

3

Determinewhetherthegraphsisstronglyconnectedorweaklyconnected.【圖片】

參考答案:

stronglyconnected

Determinewhethereachofthesegraphsisstronglyconnectedorweaklyconnected.【圖片】

參考答案:

weaklyconnected

Whatdoesthecriticalprobability【圖片】mean?

參考答案:

Itistheprobabilitythatdetermineswhetheranetworkwillhavesomeproperty.

Givenarandomnetworkof50nodesandaveragenodedegreek=10,whichofthefollowingislikelytobeclosesttotheaveragepathlengthofthatnetwok?

參考答案:

2

Considerthestep-by-stepprocessfromthepreviousquestion.Ateachstepintheprocess,supposeyourecordthesizeofthelargestcomponent.Whichofthefollowingtendstobetrueaboutthesequenceoflargestcomponentsizesasyouaddedges?

參考答案:

Increasesslowlyuntilsomethreshold,increasesveryquicklyforashorttime,andthenincreasesslowlyafterwards

Considerthefollowingprocess.First,startwithacollectionofNnodesandnolinks.Thisisclearlyadisconnectednetwork.Then,onebyone,addalinkbetweentwonodesnotalreadyconnectedtoeachother.Continueuntilyouhaveacompletenetwork.Howmanystepsareinthisprocess?

參考答案:

N-1

Whatistheruleofclustering?

參考答案:

Minimizethesimilaritybetweenclustersandmaximizethesimilaritywithinclusters

Nowwehaveadendrogramofournetwork.Howmanyclusterswouldyouretainifyoucutthedendrogramatthedistance6?【圖片】

參考答案:

5

Inarandomnetwork,thedegreedistributionofnodesalwaysfollows_____.

參考答案:

poissondistribution

Inarealnetwork,itusuallyfollows______.

參考答案:

powerlaw

Considertheexampleinthegraphbelow.Accordingtothelinearthresholdmodel,will