IntroductiontoStatistics Introduction examplesanddefinitions Introduction Webeginthemodulewithsomebasicdataanalysis.SinceStatisticsinvolves thecollectionand interpretationofdata wemustfirstknowhowto beforeundertakingamoresophisticatedanalysis. social sciences.Forexample duringthismodulewewillconsiderexamples fromBiology Medicine Agriculture Economics BusinessandMeteorology Examples thathewillsurviveforatleast5years.Bycollectingdataonsurvival
ratesofpeopleinasimilarsituation it ispossibletoobtainanempirical estimateofsurvivalrates.Wecannotknowwhetherornotthepatientwill wecanestimatetheproportionofpatientswhosurvivefromdata. Carmaintenance:Whenbuyingacertaintypeofnewcar itwouldbeuseful toknowhowmuchitisgoingtocosttorunoverthefirstthreeyearsfrom new.Ofcourse wecannotpredictexactlywhat thiswillbe-it willvary carswillgivesomeideaofthedistributionofcostsacrossthepopulation ofcarbuyers whichinturnwillprovideinformationaboutthelikelycost ofrunningthecar. Definitions Thequantitiesmeasuredinastudyarecalledrandomvariables anda particularouteiscalledanobservation.Severalobservationsare
collectivelyknownasdata.Thecollectionofallpossibleoutesiscalled thepopulation. Inpractice wecannotusuallyobservethewholepopulation.Insteadwe observeasub-setof thepopulation knownasasample.Inordertoensure thatthesamplewetakeisrepresentativeofthewholepopulation weusually likelytobeselectedforinclusioninthesample.Forexample ifweare interestedinconductingasurveyoftheamountofphysicalexercise gymnasiumwouldprovide a biasedsampleof thepopulation and theresults obtainedwouldnotgeneralisetothepopulationatlarge. Variablesareeitherqualitativeorquantitative.Qualitativevariableshave variableshavenumericoutes.Forexample survivaltime height age， numberofchildren andnumberoffaultsareallquantitativevariables.
Quantitativevariablescanbediscreteorcontinuous.Discreterandom variableshaveouteswhichcantakeonlyacountablenumberofpossible values.Thesepossiblevaluesareusuallytakentobeintegers butdon'thave randomvariableswhichtakeonlyintegervalues butyourscoreinaquiz arecontinuousrandomvariables.Often continuousrandomvariablesare roundedtothenearestinteger butthearestillconsideredtobecontinuous variablesifthereisanunderlyingcontinuousscale.Ageisagoodexampleof this. Datapresentation Introduction Asetofdataonitsownisveryhardtointerpret.Thereislotsofinformation containedinthedata butitishardtosee.Weneedwaysofunderstanding importantfeaturesofthedata andtosummariseitinmeaningfulways.
Theuseofgraphsandsummarystatisticsforunderstandingdataisan itisusefulforunderstandingthemainfeaturesofthedata fordetecting outliers anddatawhichhasbeenrecordedincorrectly.Outliersareextreme Thepresenceofoutlierscanseriouslydistortsomeofthemoreformal statisticaltechniquestobeexaminedinthesecondsemester andso preliminarydetectionandcorrectionoracmodationofsuchobservations iscrucial beforefurtheranalysistakesplace. Frequencytables It isimportanttoinvestigatetheshapeofthedistributionofarandomvariable. frequencytableshowsatallyofthenumberofdataobservationsindifferent categories. Forqualitativeanddiscretequantitativedata weoftenusealloftheobserved
统计学导论、例子和定义(英文版课件).pdf
