教程-R Programming 2nd Edition
MP4|Video:AVC960x540|Audio:AAC48KHz2ch|Duration:15H11M|
Genre:eLearning|Language:English

15+HoursofVideoInstruction

RProgrammingLiveLessons,2ndEdition,isatourthroughthemostimportantpartsofR,thestatisticalprogramminglanguage,fromtheverybasicstocomplexmodeling.Itcoversreadingdata,programmingbasics,visualization,datamunging,regression,classification,clustering,modernmachinelearning,networkanalysis,webgraphics,andtechniquesfordealingwithlargedata,bothinmemoryandindatabases.

Description

This15-hourvideoteachesyouhowtoprograminRevenifyouareunfamiliarwithstatisticaltechniques.ItstartswiththebasicsofusingRandprogressesintodatamanipulationandmodelbuilding.Userslearnthroughhands-onpracticewiththecodeandtechniques.Newmaterialcoverschainingcommands,fasterdatamanipulation,newwaystoreadrectangulardataintoR,testingcode,andthehotpackageShiny.

BasedonacourseonRandBigDatataughtbytheauthoratColumbia
DesignedfromthegrounduptohelpviewersquicklyovercomeR’slearningcurve
Packedwithhands-onpracticeopportunitiesandrealistic,downloadablecodeexamples
Presentedbyanauthorwithunsurpassedexperienceteachingstatisticalprogrammingandmodelingtonovices
ForeverypotentialRuser:programmers,datascientists,DBAs,marketers,quants,scientists,policymakers,andmanyothers
AbouttheInstructor

JaredP.LanderistheChiefDataScientistofLanderAnalytics,theorganizeroftheNewYorkOpenStatisticalProgrammingMeetup(formerlytheRMeetup)andanadjunctprofessorofStatisticsatColumbiaUniversity.WithamastersfromColumbiaUniversityinstatisticsandabachelorsfromMuhlenbergCollegeinmathematics,hehasexperienceinbothacademicresearchandindustry.Hespecializesindatamanagement,multilevelmodels,machinelearning,generalizedlinearmodels,datamanagement,visualization,andstatisticalcomputing.HeistheauthorofRforEveryone,abookaboutRProgramminggearedtowarddatascientistsandnon-statisticiansalike.Veryactiveinthedatacommunity,Jaredisafrequentspeakeratconferences,universities,andmeetupsaroundtheworld.HeisamemberoftheStrataNewYorkselectioncommittee.

SkillLevel

Beginner
Intermediate
Advanced
WhatYouWillLearn
InstallingR
Basicmath
Workingwithvariablesanddifferentdatatypes
Matrixalgebra
data.frames
Readingdata
Dataaggregationandmanipulation
plyr
dplyr
Makingstatisticalgraphs
Manipulatetext
Automaticallygeneratereportsandslideshows
DisplaydatawithpopularJavaScriptlibraries
BuildShinydashboards
BuildRpackages
IncorporateC++forfastercode
Basicstatistics
Linearmodels
Generalizedlinearmodels
Modelvalidation
Decisiontrees
Randomforests
Bootstrap
Timeseriesanalysis
Clustering
Networkanalysis
Automaticparametertuning
BayesianregressionusingStan
WhoShouldTakeThisCourse

Part1ofthelessonsisgearedtowardpeoplewhoarenewtoeitherRorprogrammingingeneral.
Part2isforRprogrammerswhoalreadyhaveanintermediatelevelofknowledgesuchasthatgainedfromReadingRforEveryoneorfromviewingPart1.

CourseRequirements
BasicProgrammingSkills
TableofContents

Part1:RasaTool

Lesson1.GettingStartedwithR
Rcanonlybeusedafterinstallation,whichfortunatelyisjustassimpleasinstallinganyotherprogram.Inthislesson,youlearnaboutwheretodownloadR,howtodecideonthebestversion,howtoinstallit,andyougetfamiliarwithitsenvironment,usingRStudioasafrontend.Wealsotakealookatthepackagesystem.

Lesson2.TheBasicBuildingBlocksinR
Risaflexibleandrobustprogramminglanguage,andusingitrequiresunderstandinghowithandlesdata.WelearnaboutperformingbasicmathinR,storingvarioustypesofdatainvariablessuchasnumeric,integer,character,andtime-basedandcallingfunctionsonthedata.

Lesson3.AdvancedDataStructuresinR
Likemanyotherlanguages,Roffersmorecomplexstoragemechanismssuchasvectors,arrays,matrices,andlists.Wetakealookatthoseandthedata.frame,aspecialstoragetypethatstronglyresemblesaspreadsheetandispartofwhatmakesworkingwithdatainRsuchapleasure.

Lesson4.ReadingDataintoR
Dataisabundantintheworld,soanalyzingitisjustamatterofgettingthedataintoR.Therearemanywaysofdoingso,themostcommonbeingreadingfromaCSVfileordatabase.Wecoverthesetechniques,andalsoimportingfromotherstatisticaltools,scrapingwebsites,andreadingExcelfiles.

Lesson5.MakingStatisticalGraphs
Visualizingdataisacrucialpartofdatasciencebothinthediscoveryphaseandwhenreportingresults.Rhaslongbeenknownforitscapabilitytoproducecompellingplots,andHadleyWickham’sggplot2packagemakesiteveneasiertoproducebetterlookinggraphics.Wecoverhistogramsscatterplots,boxplots,linecharts,andmore,inbothbasegraphicsandggplot2andthenexplorenewerpackagesggvisandrCharts.

Lesson6.BasicsofProgramming
Rhasallthestandardcomponentsofaprogramminglanguagesuchaswritingfunctions,ifstatementsandloops,allwiththeirowncaveatsandquirks.Westartwiththerequisite“Hello,World!”functionandlearnaboutargumentstofunctions,theregularifstatementandthevectorizedversion,andhowtobuildloopsandwhytheyshouldbeavoided.

Lesson7.DataMunging
Datascientistsoftenbemoanthat80%oftheirworkismanipulatingdata.Assuch,Rhasmanytoolsforthis,whichare,contrarytowhatPythonusersmaysay,easytouse.WeseehowRexcelsatgroupoperationsusingapply,lapply,andtheplyrpackage.Wealsotakealookatitsfacilitiesforjoining,combining,andrearrangingdata.Thenwespeedthatupwithtidyr,data.table,anddplyr.

Lesson8.In-Depthwithdplyr
dplyrhasbecomesuchanindispensibletool,nearlysupersedingplyr,thatitisworthdevotingextraattentionto.Soweexamineitsselect,filter,mutate,group_byandsummarizefunctions,amongothers.

Lesson9.ManipulatingStrings
Textdataisbecomingmorepervasiveintheworld,andfortunately,Rprovideswaysforbothcombiningtextandrippingitapart,whichwewalkthrough.WealsoexamineR’sextensiveregularexpressioncapabilities.

Lesson10.ReportsandSlideshowswithknitr
Successfullydeliveringtheresultsofananalysiscanbejustasimportantastheanalysisitself,soitisimportanttocommunicatetheminaneffectiveway.Inthislesson,welearnhowtouseknitrandrmarkdowntowritebothstaticandinteractiveresultsintheformofpdfdocuments,websites,HTML5slideshows,andevenWorddocuments.

Lesson11.IncludeHTMLWidgetsinHTMLDocuments
RecentyearshaveseentheadvanceofJavaScript-powereddisplaysofinformation,andthehtmlwidgetspackageenablesRtotakeadvantageofarbitraryJavaScriptlibraries.Inparticular,welookatdatatableforatabulardisplayofdata,bokehforrichwebplots,andleafletforpowerfulmapping.

Lesson12.Shiny
BuiltbyRstudio,ShinyisatoolforbuildinginteractivedatadisplaysanddashboardsallwithinR.ThisallowstheRprogrammertoconveyresultsinacompelling,user-richexperienceinalanguageheorsheisfamiliarwith.

Lesson13.PackageBuilding
BuildingpackagesisagreatwaytocontributebacktotheRcommunity,anddoingsohasneverbeeneasierthankstoHadleyWickham'sdevtoolspackage.Thislessoncoversalltherequirementsforapackageandhowtogoaboutauthoring,testing,anddistributingthem.

Lesson14.RcppforFasterCode
SometimespureRcodeisnotfastenough,andextraspeedisrequired.RcppenablesRprogrammerstoseamlesslyintegrateC++codeintotheirRcode.Wegooverthebasicsofgettingthetwolanguagesworkingtogether,writesomespeedyfunctionsinC++,andevenintegrateC++intoRpackages.

Part2:RforStatistics,Modeling,andMachineLearning

Lesson15.BasicStatistics
Naturally,Rhasallthebasicswhenitcomestostatisticssuchasmeans,variance,correlation,t-tests,andANOVAs.Welookatallthedifferentwaysthosecanbecomputed.

Lesson16.LinearModels
Theworkhorseofstatisticsisregressionanditsextensions.Thisconsistsoflinearmodels,generalizedlinearmodels—includinglogisticandPoissonregression&mdashandsurvivalmodels.WelookathowtofitthesemodelsinRandhowtoevaluatethemusingmeasuressuchasmeansquarederror,deviance,andAIC.

Lesson17.OtherModels
Beyondregressiontherearemanyothertypesofmodelsthatcanbefittodata.Modelscoveredincluderegularizationwiththeelasticnet,Bayesianshrinkage,nonlinearmodelssuchasnonlinearleastsquares,splinesandgeneralizedadditivemodels,decisiontress,andrandomforests.

Lesson18.TimeSeries
Specialcaremustbetakenwithdatawherethereistime-basedcorrelation,otherwiseknownasautocorrelation.WelookatsomecommonmethodsfordealingwithtimeseriessuchasARIMA,VAR,andGARCH.

Lesson19.Clustering
Afocalpointofmodernmachinelearningisclustering,thepartitioningofdataintogroups.Weexplorethreepopularmethods:K-means,K-medoids,andhierarchicalclustering.

Lesson20.MoreMachineLearning
Twoareasseeingincreasinginterestarerecommendationenginesandtextmining,whichweillustratewithRecommenderLab,RTextTools,andtheirlbapackageforfastmatrixfactorization.

Lesson21.NetworkAnalysis
Theworldisrichwithnetworkdatathatarenicelystudiedwithgraphicalmodels.Weshowyouhowtoanalyzeandvisualizenetworksusingtheigraphpackage.

Lesson22.AutomaticParameterTuningwithCaret
Machinelearningmodelsoftenhaveparametersthatneedtuning,whichcansignificantlyaffectthequalityofthemodel.TheCaretpackage,byMaxKuhn,makesfindingoptimalparametervalueseasytofind.

Lesson23.FitaBayesianModelwithRStan
Bayesiandataanalysisusessimulationstofitbothsimpleandcomplexmodels.AndrewGelman’snewlanguage,Stan,makesthisfasterandeasierthaneverbefore.Weexplorethisbyfittingasimplelinearregressionandvarying-interceptmultilevelmodel.