Modern Statistics: A Computer-Based Approach with Python (Statistics for Industry, Technology, and Engineering)

452 Pages·2022·6.559 MB·English

by  Ron S. Kenett, Shelemyahu Zacks, Peter Gedeck

Statistics for Industry, Technology, and
Engineering
SeriesEditor
DavidSteinberg,TelAvivUniversity,TelAviv,Israel
EditorialBoardMembers
V.RoshanJoseph,GeorgiaInstituteofTechnology,Atlanta,GA,USA
RonS.Kenett,KPALtd.RaananaandSamuelNeamanInstitute,Technion,
Haifa,Israel
ChristineAnderson-Cook,LosAlamosNationalLaboratory,LosAlamos,USA
BradleyJones,SASInstitute,JMPDivision,Cary,USA
FugeeTsung,HongKongUniversityofScienceandTechnology,HongKong,
HongKong
The Statistics for Industry, Technology, and Engineering series will present up-to-
date statistical ideas and methods that are relevant to researchers and accessible
to an interdisciplinary audience: carefully organized authoritative presentations,
numerousillustrativeexamplesbasedoncurrentpractice,reliablemethods,realistic
data sets, and discussions of select new emerging methods and their application
potential.Publicationswillappealtoabroadinterdisciplinaryreadershipincluding
both researchers and practitioners in applied statistics, data science, industrial
statistics, engineering statistics, quality control, manufacturing, applied reliability,
andgeneralqualityimprovementmethods.
PrincipalTopicAreas:
* Quality Monitoring * Engineering Statistics * Data Analytics * Data Science *
TimeSerieswithApplications*SystemsAnalyticsandControl*Stochasticsand
Simulation * Reliability * Risk Analysis * Uncertainty Quantification * Decision
Theory * Survival Analysis * Prediction and Tolerance Analysis * Multivariate
Statistical Methods * Nondestructive Testing * Accelerated Testing * Signal
Processing*ExperimentalDesign*SoftwareReliability*NeuralNetworks*
Theserieswillincludeprofessionalexpositorymonographs,advancedtextbooks,
handbooks, general references,thematic compilations of applications/case studies,
andcarefullyeditedsurveybooks.
Ron S. Kenett • Shelemyahu Zacks • Peter Gedeck
Modern Statistics
A Computer-Based Approach with Python
RonS.Kenett ShelemyahuZacks
KPALtd.RaananaandSamuelNeaman MathematicalSciences
Institute,Technion BinghamtonUniversity
Haifa,Israel McLean,VA,USA
PeterGedeck
DataScience
UniversityofVirginia
FallsChurch,VA,USA
Thisworkcontainsmediaenhancements,whicharedisplayedwitha“play”icon.Materialintheprint
bookcanbeviewedonamobiledevicebydownloadingtheSpringerNature“MoreMedia”appavailable
inthemajorappstores.Themediaenhancementsintheonlineversionoftheworkcanbeaccessed
directlybyauthorizedusers.
ISSN2662-5555 ISSN2662-5563 (electronic)
StatisticsforIndustry,Technology,andEngineering
ISBN978-3-031-07565-0 ISBN978-3-031-07566-7 (eBook)
https://doi.org/10.1007/978-3-031-07566-7
MathematicsSubjectClassification:62E15,62G30,62M10,62P30,62P10,97K40,97K70,97K80
©TheEditor(s)(ifapplicable)andTheAuthor(s),underexclusivelicensetoSpringerNatureSwitzerland
AG2022
Thisworkissubjecttocopyright.AllrightsaresolelyandexclusivelylicensedbythePublisher,whether
thewholeorpartofthematerialisconcerned,specificallytherightsoftranslation,reprinting,reuse
ofillustrations,recitation,broadcasting,reproductiononmicrofilmsorinanyotherphysicalway,and
transmissionorinformationstorageandretrieval,electronicadaptation,computersoftware,orbysimilar
ordissimilarmethodologynowknownorhereafterdeveloped.
Theuseofgeneraldescriptivenames,registerednames,trademarks,servicemarks,etc.inthispublication
doesnotimply,evenintheabsenceofaspecificstatement,thatsuchnamesareexemptfromtherelevant
protectivelawsandregulationsandthereforefreeforgeneraluse.
Thepublisher,theauthors,andtheeditorsaresafetoassumethattheadviceandinformationinthisbook
arebelievedtobetrueandaccurateatthedateofpublication.Neitherthepublishernortheauthorsor
theeditorsgiveawarranty,expressedorimplied,withrespecttothematerialcontainedhereinorforany
errorsoromissionsthatmayhavebeenmade.Thepublisherremainsneutralwithregardtojurisdictional
claimsinpublishedmapsandinstitutionalaffiliations.
This book is published under the imprint Birkhäuser, www.birkhauser-science.com by the registered
companySpringerNatureSwitzerlandAG
Theregisteredcompanyaddressis:Gewerbestrasse11,6330Cham,Switzerland
TomywifeSima,ourchildrenandtheir
children:Yonatan,Alma,Tomer,Yadin,Aviv,
Gili,Matan,Eden,andEthan.RSK
TomywifeHanna,oursonsYuvalandDavid,
andtheirfamilieswithlove.SZ
ToJanetwithlove.PG
Preface
Statisticshasdevelopedbycombiningtheneedsofscience,business,industry,and
government. More recent development is connected with methods for generating
insightsfromdata,usingstatisticaltheoryanddeliveryplatforms.Thisintegration
isatthecoreofappliedstatisticsandmostoftheoreticalstatistics.
Beforethebeginningofthetwentiethcentury,statisticsmeantobserveddataand
descriptivesummaryfigures,suchasmeans,variances,indices,etc.,computedfrom
data.Withtheintroductionoftheχ2-testforgoodnessoffitbyKarlPearson(1900)
and the t-test by Gosset (Student, 1908) for drawing inference on the mean of a
normal population, statistics became a methodology of analyzing sample data to
determinethevalidityofhypotheses aboutthesourceofthedata(thepopulation).
Fisher (1922) laid the foundations for statistics as a discipline. He considered the
objectofstatisticalmethodstobereducingdataintotheessentialstatistics,andhe
identifiedthreeproblemsthatariseindoingso:
1. Specification-choosingtherightmathematicalmodelforapopulation
2. Estimation-methodstocalculate,fromasample,estimatesoftheparametersof
thehypotheticalpopulation
3. Distribution-propertiesofstatisticsderivedfromsamples
Fortyyearslater,Tukey(1962)envisionedadata-centricdevelopmentofstatis-
tics,sketchingthepathwaytodatascience.Fortyyearsafterthat,weenteredtheage
of big data, data science, artificial intelligence, and machine learning. These new
developmentsarebuiltonthemethods,applications,andexperienceofstatisticians
aroundtheworld.
The first two authors started collaborating on a book in the early 1990s. In
1998,wepublishedwithDuxburyWadsworthModernIndustrialStatistics:Design
and Control of Quality and Reliability. The book appeared in a Spanish edition
(Estadística Industrial Moderna: Diseño y Control de Calidad y la Confiabilidad,
Thomson International, 2000). An abbreviated edition was published as Modern
Statistics: A Computer based Approach (Thomson Learning, 2001); this was
followedbyaChineseedition(ChinaStatisticsPress,2003)andasoftcoveredition,
(Brooks/Cole, 2004). The book used QuickBasic, S-Plus, and MINITAB. In 2014
vii
viii Preface
we published, with Wiley, an extended second edition titled Modern Industrial
Statistics: With Applications in R, MINITAB and JMP. That book was translated
into Vietnamese by the Vietnam Institute for Advanced Studies in Mathematics
(VIASM,2016).Athird,expandededition,waspublishedbyWileyin2021.
This book is about modern statistics with Python. It reflects many years of
experience of the authors in doing research, teaching and applying statistics in
science, healthcare, business, defense, and industry domains. The book invokes
over 40 case studies and provides comprehensive Python applications. In 2019,
there were 8.2 million developers in the world who code using Python which is
consideredthefastest-growingprogramminglanguage.AspecialPythonpackage,
mistat,isavailablefordownloadhttps://gedeck.github.io/mistat-code-solutions/
ModernStatistics/. Everything in the book can be reproduced with mistat. We
therefore provide, in this book, an integration of needs, methods, and delivery
platformforalargeaudienceandawiderangeofapplications.
Modern Statistics: A Computer-Based Approach with Python is a companion
texttoanotherbookpublishedbySpringertitled:IndustrialStatistics:AComputer
BasedApproachwithPython.Bothbooksincludemutualcrossreferences,butboth
books are stand-alone publications. This book can be used as textbook in a one
semester or two semester course on modern statistics. The technical level of the
presentation in both books can serve both undergraduate and graduate students.
The example and case studies provide access to hands on teaching and learning.
Everychapterincludesexercises,datasets,andPythonapplications.Thesecanbe
used in regular classroom setups, flipped classroom setups, and online or hybrid
education programs. The companion text is focused on industrial statistics with
special chapters on advanced process monitoring methods, cybermanufacturing,
computerexperiments,andBayesianreliability.ModernStatisticsisafoundational
textandcanbecombinedwithanyprogramrequiringdataanalysisinitscurriculum.
This, for example, can be courses in data science, industrial statistics, physics,
biology, chemistry, economics, psychology, social sciences, or any engineering
discipline.
Modern Statistics: A Computer-Based Approach with Python includes eight
chapters.Chapter1isonanalyzingvariabilitywithdescriptivestatistics.Chapter2
isonprobabilitymodelsanddistributionfunctions.Chapter3introducesstatistical
inference and bootstrapping. Chapter 4 is on variability in several dimensions and
regression models. Chapter 5 covers sampling for estimation of finite population
quantities, a common situation when one wants to infer on a population from a
sample. Chapter 6 is dedicated to time series analysis and prediction. Chapters 7
and8areaboutmoderndataanalyticmethods.
Industrial Statistics: A Computer-Based Approach with Python contains 11
chapters:Chapter1—IntroductiontoIndustrialStatistics,Chapter2—BasicTools
and Principles of Process Control, Chapter 3—Advanced Methods of Statistical
Process Control,Chapter 4—Multivariate StatisticalProcess Control,Chapter 5—
Classical Design and Analysis of Experiments, Chapter 6—Quality by Design,
Chapter 7—Computer Experiments, Chapter 8—Cybermanufacturing and Digital
Twins,Chapter9—ReliabilityAnalysis,Chapter10—BayesianReliabilityEstima-
Preface ix
tion and Prediction, and Chapter 11—Sampling Plans for Batch and Sequential
Inspection. This second book is focused on industrial statistics with applications
tomonitoring,diagnostics,prognostic,andprescriptiveanalytics.Itcanbeusedas
a stand-alone book, or in conjunction with Modern Statistics. Both books include
solutionmanualstoexerciseslistedattheendofeachchapter.Thiswasdesignedto
supportself-learningaswellasinstructorledcourses.
Wemadeeverypossibleefforttoensurethecalculationsarecorrectandthetext
is clear. However, should errors have skipped to the printed version, we would
appreciate feedback from readers noticing these. In general, any feedback will be
muchappreciated.
Finally, we would like to thank the team at Springer Birkhäuser, including
DanaKnowlesandChristopherTominich.Theymadeeverythinginthepublication
processlookeasy.
Ra’anana,Israel RonS.Kenett
McLean,VA,USA ShelemyahuZacks
FallsChurch,VA,USA PeterGedeck
April2022
Contents
1 AnalyzingVariability:DescriptiveStatistics.............................. 1
1.1 RandomPhenomenaandtheStructureofObservations............. 1
1.2 AccuracyandPrecisionofMeasurements............................ 6
1.3 ThePopulationandtheSample....................................... 8
1.4 DescriptiveAnalysisofSampleValues............................... 9
1.4.1 FrequencyDistributionsofDiscreteRandomVariables ... 9
1.4.2 FrequencyDistributionsofContinuousRandom
Variables...................................................... 14
1.4.3 StatisticsoftheOrderedSample ............................ 17
1.4.4 StatisticsofLocationandDispersion....................... 19
1.5 PredictionIntervals.................................................... 23
1.6 AdditionalTechniquesofExploratoryDataAnalysis ............... 25
1.6.1 DensityPlots ................................................. 25
1.6.2 BoxandWhiskersPlots ..................................... 27
1.6.3 QuantilePlots ................................................ 29
1.6.4 Stem-and-LeafDiagrams.................................... 30
1.6.5 RobustStatisticsforLocationandDispersion.............. 31
1.7 ChapterHighlights..................................................... 34
1.8 Exercises............................................................... 34
2 ProbabilityModelsandDistributionFunctions.......................... 39
2.1 BasicProbability....................................................... 39
2.1.1 EventsandSampleSpaces:FormalPresentation
ofRandomMeasurements................................... 39
2.1.2 BasicRulesofOperationswithEvents:Unions
andIntersections ............................................. 41
2.1.3 ProbabilitiesofEvents....................................... 44
2.1.4 ProbabilityFunctionsforRandomSampling............... 46
2.1.5 ConditionalProbabilitiesandIndependenceofEvents .... 49
2.1.6 Bayes’TheoremandItsApplication........................ 51
2.2 RandomVariablesandTheirDistributions........................... 54
xi
xii Contents
2.2.1 DiscreteandContinuousDistributions...................... 55
2.2.1.1 DiscreteRandomVariables...................... 55
2.2.1.2 ContinuousRandomVariables .................. 56
2.2.2 ExpectedValuesandMomentsofDistributions............ 59
2.2.3 TheStandardDeviation,Quantiles,Measuresof
Skewness,andKurtosis...................................... 62
2.2.4 MomentGeneratingFunctions.............................. 65
2.3 FamiliesofDiscreteDistribution ..................................... 66
2.3.1 TheBinomialDistribution................................... 66
2.3.2 TheHypergeometricDistribution........................... 69
2.3.3 ThePoissonDistribution .................................... 72
2.3.4 TheGeometricandNegativeBinomialDistributions...... 74
2.4 ContinuousDistributions.............................................. 78
2.4.1 TheUniformDistributionontheInterval(a,b),
a <b ......................................................... 78
2.4.2 TheNormalandLog-NormalDistributions ................ 79
2.4.2.1 TheNormalDistribution......................... 79
2.4.2.2 TheLog-NormalDistribution ................... 84
2.4.3 TheExponentialDistribution................................ 85
2.4.4 TheGammaandWeibullDistributions..................... 88
2.4.5 TheBetaDistributions....................................... 92
2.5 Joint,Marginal,andConditionalDistributions....................... 93
2.5.1 JointandMarginalDistributions............................ 93
2.5.2 CovarianceandCorrelation.................................. 96
2.5.3 ConditionalDistributions.................................... 99
2.6 SomeMultivariateDistributions...................................... 102
2.6.1 TheMultinomialDistribution............................... 102
2.6.2 TheMulti-HypergeometricDistribution.................... 104
2.6.3 TheBivariateNormalDistribution.......................... 105
2.7 DistributionofOrderStatistics........................................ 108
2.8 LinearCombinationsofRandomVariables .......................... 111
2.9 LargeSampleApproximations........................................ 117
2.9.1 TheLawofLargeNumbers ................................. 117
2.9.2 TheCentralLimitTheorem ................................. 117
2.9.3 SomeNormalApproximations.............................. 119
2.10 AdditionalDistributionsofStatisticsofNormalSamples........... 120
2.10.1 DistributionoftheSampleVariance ........................ 121
2.10.2 The“Student”t-Statistic..................................... 122
2.10.3 DistributionoftheVarianceRatio........................... 123
2.11 ChapterHighlights..................................................... 125
2.12 Exercises............................................................... 126
3 StatisticalInferenceandBootstrapping................................... 139
3.1 SamplingCharacteristicsofEstimators .............................. 139
3.2 SomeMethodsofPointEstimation................................... 141