Professional Documents
Culture Documents
What is Statistics?
Soonhui Lee
College of Businees
HUFS
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.1
What is Statistics?
Statisticsisawaytogetinformationfromdata.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.2
What is Statistics?
Statisticsisawaytogetinformationfromdata
Statistics
Data
Information
Statisticsisatoolforcreatingnewunderstandingfromasetof
numbers.
Definitions:OxfordEnglishDictionary
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.3
1.4
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.5
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.6
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.7
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.8
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.9
Descriptive Statistics
Descriptivestatisticsdealswithmethodsoforganizing,
summarizing,andpresentingdatainaconvenientand
informativeway.
Oneformofdescriptivestatisticsusesgraphicaltechniques,
whichallowstatisticspractitionerstopresentdatainwaysthat
makeiteasyforthereadertoextractusefulinformation.
Chapter2introducesseveralgraphicalmethods.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.10
Descriptive Statistics
Anotherformofdescriptivestatisticsusesnumerical
techniquestosummarizedata.
Themeanandmedianarepopularnumericaltechniquesto
describethelocationofthedata.
Therange,variance,andstandarddeviationmeasurethe
variabilityofthedata
Chapter4introducesseveralnumericalstatisticalmeasures
thatdescribedifferentfeaturesofthedata.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.11
studentshasofferedPepsiColaanexclusivityagreementthat
wouldgivePepsiexclusiverightstosellitsproductsatall
universityfacilitiesforthenextyearwithanoptionforfuture
years.
Inreturn,theuniversitywouldreceive35%oftheoncampus
revenuesandanadditionallumpsumof$200,000peryear.
Pepsihasbeengiven2weekstorespond.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.12
Pepsicurrentlysellsanaverageof22,000cansperweek(over
the40weeksoftheyearthattheuniversityoperates).
Thecanssellforanaverageof75centseach.Thecosts
includinglaboramountto20centspercan.
Pepsiisunsureofitsmarketsharebutsuspectsitis
considerablylessthan50%.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.13
Pepsiwouldsell88,000(22,000is25%of88,000)cansper
weekor3,520,000cansperyear.
Theprofitorlosscanbecalculated.
Theonlyproblemisthatwedonotknowhowmanysoft
drinksaresoldweeklyattheuniversity.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.14
Accordingly,sheorganizesasurveythatasks500studentsto
keeptrackofthenumberofsoftdrinkstheypurchaseinthe
next7days.
Theresponsesarestoredinafileonthediskthataccompanies
thisbook.Case12.1
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.15
Inferential statistics
TheinformationwewouldliketoacquireinCase12.1isan
estimateofannualprofitsfromtheexclusivityagreement.The
dataarethenumbersofcansofsoftdrinksconsumedin7days
bythe500studentsinthesample.
Wewanttoknowthemeannumberofsoftdrinksconsumed
byall50,000studentsoncampus.
Toaccomplishthisgoalweneedanotherbranchofstatistics
inferentialstatistics.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.16
Inferential statistics
Inferentialstatisticsisabodyofmethodsusedtodraw
conclusionsorinferencesaboutcharacteristicsofpopulations
basedonsampledata.Thepopulationinquestioninthiscase
isthesoftdrinkconsumptionoftheuniversity's50,000
students.Thecostofinterviewingeachstudentwouldbe
prohibitiveandextremelytimeconsuming.Statistical
techniquesmakesuchendeavorsunnecessary.Instead,wecan
sampleamuchsmallernumberofstudents(thesamplesizeis
500)andinferfromthedatathenumberofsoftdrinks
consumedbyall50,000students.Wecanthenestimateannual
profitsforPepsi.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.17
Example 12.5
Whenanelectionforpoliticalofficetakesplace,thetelevision
networkscancelregularprogrammingandinsteadprovide
electioncoverage.
Whentheballotsarecountedtheresultsarereported.
However,forimportantofficessuchaspresidentorsenatorin
largestates,thenetworksactivelycompetetoseewhichwill
bethefirsttopredictawinner.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.18
Example 12.5
Thisisdonethroughexitpolls,whereinarandomsampleof
voterswhoexitthepollingboothisaskedforwhomthey
voted.
Fromthedatathesampleproportionofvoterssupportingthe
candidatesiscomputed.
Astatisticaltechniqueisappliedtodeterminewhetherthere
isenoughevidencetoinferthattheleadingcandidatewill
garnerenoughvotestowin.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.19
Example 12.5
TheexitpollresultsfromthestateofFloridaduringthe2000
yearelectionswererecorded(onlythevotesoftheRepublican
candidateGeorgeW.BushandtheDemocratAlbertGore).
Supposethattheresults(765peoplewhovotedforeitherBush
orGore)werestoredonafileonthedisk.(1=Goreand2=
Bush)
Xm1205
Thenetworkanalystswouldliketoknowwhethertheycan
concludethatGeorgeW.BushwillwinthestateofFlorida.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.20
Example 12.5
Example12.5describesaverycommonapplicationof
statisticalinference.
Thepopulationthetelevisionnetworkswantedtomake
inferencesaboutistheapproximately5millionFloridianswho
votedforBushorGoreforpresident.
Thesampleconsistedofthe765peoplerandomlyselectedby
thepollingcompanywhovotedforeitherofthetwomain
candidates.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.21
Example 12.5
Thecharacteristicofthepopulationthatwewouldliketo
knowistheproportionofthetotalelectoratethatvotedfor
Bush.
Specifically,wewouldliketoknowwhethermorethan50%
oftheelectoratevotedforBush(countingonlythosewho
votedforeithertheRepublicanorDemocraticcandidate).
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.22
Example 12.5
Becausewewillnotaskeveryoneofthe5millionactual
votersforwhomtheyvoted,wecannotpredicttheoutcome
with100%certainty.
Asamplethatisonlyasmallfractionofthesizeofthe
populationcanleadtocorrectinferencesonlyacertain
percentageofthetime.
Youwillfindthatstatisticspractitionerscancontrolthat
fractionandusuallysetitbetween90%and99%.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.23
Sample
Asampleisasetofdatadrawnfromthe
population.
Potentiallyverylarge,butlessthanthepopulation.
E.g.asampleof765votersexitpolledonelectionday.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.24
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.25
Sample
Subset
Parameter
Statistic
PopulationshaveParameters,
SampleshaveStatistics.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.26
Descriptive Statistics
aremethodsoforganizing,summarizing,andpresenting
datainaconvenientandinformativeway.Thesemethods
include:
GraphicalTechniques(Chapter2),and
NumericalTechniques(Chapter4).
Theactualmethoduseddependsonwhatinformationwe
wouldliketoextract.Areweinterestedin
measure(s)ofcentrallocation?and/or
measure(s)ofvariability(dispersion)?
DescriptiveStatisticshelpstoanswerthesequestions
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.27
Inferential Statistics
DescriptiveStatisticsdescribethedatasetthatsbeing
analyzed,butdoesntallowustodrawanyconclusionsor
makeanyinterferencesaboutthedata.Henceweneed
anotherbranchofstatistics:inferentialstatistics.
Inferentialstatisticsisalsoasetofmethods,butitisusedto
drawconclusionsorinferencesaboutcharacteristicsof
populationsbasedondatafromasample.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.28
Statistical Inference
Statisticalinferenceistheprocessofmakinganestimate,
prediction,ordecisionaboutapopulationbasedonasample.
Population
Sample
Inference
Statistic
Parameter
WhatcanweinferaboutaPopulationsParameters
basedonaSamplesStatistics?
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.29
Statistical Inference
Weusestatisticstomakeinferencesaboutparameters.
Therefore,wecanmakeanestimate,prediction,ordecision
aboutapopulationbasedonsampledata.
Thus,wecanapplywhatweknowaboutasampletothe
largerpopulationfromwhichitwasdrawn!
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.30
Statistical Inference
Rationale:
Largepopulationsmakeinvestigatingeachmemberimpractical
andexpensive.
Easierandcheapertotakeasampleandmakeestimatesaboutthe
populationfromthesample.
However:
Suchconclusionsandestimatesarenotalwaysgoingtobecorrect.
Forthisreason,webuildintothestatisticalinferencemeasuresof
reliability,namelyconfidencelevelandsignificancelevel.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.31
E.g.aconfidencelevelof95%meansthat,estimatesbasedonthis
formofstatisticalinferencewillbecorrect95%ofthetime.
Whenthepurposeofthestatisticalinferenceistodrawa
conclusionaboutapopulation,thesignificancelevel
measureshowfrequentlytheconclusionwillbewronginthe
longrun.
E.g.a5%significancelevelmeansthat,inthelongrun,thistype
ofconclusionwillbewrong5%ofthetime.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.32
Thisrelationshipcanalsobestatedas:
ConfidenceLevel
+SignificanceLevel
=1
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.33
Inthiscase,ourconfidencelevelis95%(19/20=0.95),
whileoursignificancelevelis5%.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.34
Statistical Applications in
Business
Statisticalanalysisplaysanimportantroleinvirtuallyall
aspectsofbusinessandeconomics.
Throughoutthiscourse,wewillseeapplicationsofstatistics
inaccounting,economics,finance,humanresources
management,marketing,andoperationsmanagement.
2015CengageLearning.AllRightsReserved.Maynotbecopied,scanned,orduplicated,inwholeorinpart,exceptforuseaspermittedina
licensedistributedwithacertainproductorserviceorotherwiseonapasswordprotectedwebsiteforclassroomuse.
1.35