版权说明:本文档由用户提供并上传,收益归属内容提供方,若内容存在侵权,请进行举报或认领
文档简介
Chi-SquareTestsChapter11ObjectivesInthischapter,youlearn:Howandwhentousethechi-squaretestforcontingencytablesContingencyTablesContingencyTablesUsefulinsituationscomparingmultiplepopulationproportionsUsedtoclassifysampleobservationsaccordingtotwoormorecharacteristicsAlsocalledacross-classificationtable.DCOVA Left-Handedvs.Gender
DominantHand:Leftvs.Right Gender:Malevs.Female2categoriesforeachvariable,sothisiscalleda2x2tableSupposeweexamineasampleof300childrenContingencyTableExampleDCOVAContingencyTableExampleSampleresultsorganizedinacontingencytable:(continued)GenderHandPreferenceLeftRightFemale12108120Male2415618036264300120Females,12werelefthanded180Males,24werelefthandedsamplesize=n=300:DCOVA
2TestfortheDifferenceBetweenTwoProportionsIfH0istrue,thentheproportionofleft-handedfemalesshouldbethesameastheproportionofleft-handedmalesThetwoproportionsaboveshouldbethesameastheproportionofleft-handedpeopleoverallH0:π1=π2(Proportionoffemaleswhoareleft handedisequaltotheproportionof maleswhoarelefthanded)H1:π1≠π2(Thetwoproportionsarenotthesame)DCOVATheChi-SquareTestStatisticwhere:
fo=observedfrequencyinaparticularcell
fe=expectedfrequencyinaparticularcellifH0istrue
(Assumed:eachcellinthecontingencytablehasexpectedfrequencyofatleast5)TheChi-squareteststatisticis:DCOVADecisionRule
2
2αDecisionRule:If,rejectH0,otherwise,donotrejectH0Theteststatisticapproximatelyfollowsachi-squareddistributionwithonedegreeoffreedom0
RejectH0DonotrejectH0DCOVAComputingthe
OverallProportionHere:
120Females,12werelefthanded180Males,24werelefthandedi.e.,basedonall300childrentheproportionoflefthandersis0.12,thatis,12% Theoverallproportionis:DCOVAFindingExpectedFrequenciesToobtaintheexpectedfrequencyforlefthandedfemales,multiplytheaverageproportionlefthanded(p)bythetotalnumberoffemalesToobtaintheexpectedfrequencyforlefthandedmales,multiplytheaverageproportionlefthanded
(p)bythetotalnumberofmalesIfthetwoproportionsareequal,then
P(LeftHanded|Female)=P(LeftHanded|Male)=.12i.e.,wewouldexpect (.12)(120)=14.4femalestobelefthanded (.12)(180)=21.6malestobelefthandedDCOVAObservedvs.ExpectedFrequenciesGenderHandPreferenceLeftRightFemaleObserved=12Expected=14.4Observed=108Expected=105.6120MaleObserved=24Expected=21.6Observed=156Expected=158.418036264300DCOVAGenderHandPreferenceLeftRightFemaleObserved=12Expected=14.4Observed=108Expected=105.6120MaleObserved=24Expected=21.6Observed=156Expected=158.418036264300TheChi-SquareTestStatisticTheteststatisticis:DCOVADecisionRuleDecisionRule:If>3.841,rejectH0,otherwise,donotrejectH0Here,=0.7576<=3.841,sowedonotrejectH0andconcludethatthereisnotsufficientevidencethatthetwoproportionsaredifferentat=0.05
2
20.05=3.8410
0.05RejectH0DonotrejectH0DCOVAExtendthe
2testtothecasewithmorethantwoindependentpopulations:
2TestforDifferencesAmong
MoreThanTwoProportionsH0:π1=π2=…=πcH1:Notalloftheπjareequal(j=1,2,…,c)DCOVATheChi-SquareTestStatisticWhere:
fo=observedfrequencyinaparticularcellofthe2xctable
fe=expectedfrequencyinaparticularcellifH0istrue
(Assumed:eachcellinthecontingencytablehasexpectedfrequencyofatleast1)TheChi-squareteststatisticis:DCOVAComputingthe
OverallProportionTheoverallproportionis:Expectedcellfrequenciesfortheccategoriesarecalculatedasinthe2x2case,andthedecisionruleisthesame:Whereisfromthechi-squareddistributionwithc–1degreesoffreedomDecisionRule:If,rejectH0,otherwise,donotrejectH0DCOVA
2TestofIndependenceSimilartothe
2testforequalityofmorethantwoproportions,butextendstheconcepttocontingencytableswithrrowsandccolumnsH0:Thetwocategoricalvariablesareindependent (i.e.,thereisnorelationshipbetweenthem)H1:Thetwocategoricalvariablesaredependent (i.e.,thereisarelationshipbetweenthem)DCOVA
2TestofIndependencewhere:
fo=observedfrequencyinaparticularcelloftherxctable
fe=expectedfrequencyinaparticularcellifH0istrue
(Assumed:eachcellinthecontingencytablehasexpectedfrequencyofatleast1)TheChi-squareteststatisticis:(continued)DCOVAExpectedCellFrequenciesExpectedcellfrequencies:Where: rowtotal=sumofallfrequenciesintherow columntotal=sumofallfrequenciesinthecolumn n=overallsamplesizeDCOVADecisionRuleThedecisionruleisWhereisfromthechi-squaredistributionwith(r–1)(c–1)degreesoffreedomIf,rejectH0,otherwise,donotrejectH0DCOVAExampleThemealplanselectedby200studentsisshownbelow:ClassStandingNumberofmealsperweekTotal20/week10/weeknoneFresh.24321470Soph.22261260Junior1014630Senior14161040Total708842200DCOVAExampleThehypothesistobetestedis:(continued)H0:Mealplanandclassstandingareindependent (i.e.,thereisnorelationshipbetweenthem)H1:Mealplanandclassstandingaredependent (i.e.,thereisarelationshipbetweenthem)DCOVAClassStandingNumberofmealsperweekTotal20/wk10/wknoneFresh.24321470Soph.22261260Junior1014630Senior14161040Total708842200ClassStandingNumberofmealsperweekTotal20/wk10/wknoneFresh.24.530.814.770Soph.21.026.412.660Junior10.513.26.330Senior14.017.68.440Total708842200Observed:ExpectedcellfrequenciesifH0istrue:Exampleforonecell:Example:
ExpectedCellFrequencies(continued)DCOVAExample:TheTestStatisticTheteststatisticvalueis:(continued)=12.592fromthechi-squaredistributionwith(4–1)(3–1)=6degreesoffreedomDCOVAExample:
DecisionandInterpretation(continued)DecisionRule:If>12.592,rejectH0,otherwise,donotrejectH0Here,=0.709<=12.592,sodonotrejectH0
Conclusion:thereisnotsufficientevidencethatmealplanandclassstandingarerelatedat=0.05
2
20.05=12.5920
0.05RejectH0DonotrejectH0DCOVAChapterSummaryInthischapterwediscussed:Howandwhentousethechi-squaretestforcontingencytablesIntroductiontoMultipleRegressionChapter13ObjectivesInthischapter,youlearn:
HowtodevelopamultipleregressionmodelHowtointerprettheregressioncoefficientsHowtodeterminewhichindependentvariablestoincludeintheregressionmodelHowtousecategoricalindependentvariablesinaregressionmodelTheMultipleRegressionModelIdea:Examinethelinearrelationshipbetween1dependent(Y)&2ormoreindependentvariables(Xi)MultipleRegressionModelwithkIndependentVariables:Y-interceptPopulationslopesRandomErrorDCOVAMultipleRegressionEquationThecoefficientsofthemultipleregressionmodelareestimatedusingsampledataEstimated(orpredicted)valueofYEstimatedslopecoefficientsMultipleregressionequationwithkindependentvariables:EstimatedinterceptInthischapterwewilluseExcelandMinitabtoobtaintheregressionslopecoefficientsandotherregressionsummarymeasures.DCOVATwovariablemodelYX1X2SlopeforvariableX1SlopeforvariableX2MultipleRegressionEquation(continued)DCOVAAdistributoroffrozendessertpieswantstoevaluatefactorsthoughttoinfluencedemandDependentvariable:Piesales(unitsperweek)Independentvariables:Price(in$)
Advertising($100’s)Dataarecollectedfor15weeksExample:
2IndependentVariablesDCOVAPieSalesExampleSales=b0+b1(Price) +b2(Advertising)WeekPieSalesPrice($)Advertising($100s)13505.503.324607.503.333508.003.044308.004.553506.803.063807.504.074304.503.084706.403.794507.003.5104905.004.0113407.203.5123007.903.2134405.904.0144505.003.5153007.002.7Multipleregressionequation:DCOVAExcelMultipleRegressionOutputRegressionStatisticsMultipleR0.72213RSquare0.52148AdjustedRSquare0.44172StandardError47.46341Observations15ANOVA
dfSSMSFSignificanceFRegression229460.02714730.0136.538610.01201Residual1227033.3062252.776Total1456493.333
CoefficientsStandardErrortStatP-valueLower95%Upper95%Intercept306.52619114.253892.682850.0199357.58835555.46404Price-24.9750910.83213-2.305650.03979-48.57626-1.37392Advertising74.1309625.967322.854780.0144917.55303130.70888DCOVAMinitabMultipleRegressionOutputTheregressionequationisSales=307-25.0Price+74.1Advertising
Predictor
Coef
SECoef
T
PConstant 306.50
114.302.68
0.020Price -24.98
10.83
-2.31
0.040Advertising 74.13
25.97
2.85
0.014
S=47.4634
R-Sq=52.1%
R-Sq(adj)=44.2%
AnalysisofVariance
Source
DF
SS
MS
F
PRegression
2
29460
14730
6.54
0.012ResidualError
12
27033
2253Total
14
56493DCOVATheMultipleRegressionEquationb1=-24.975:saleswilldecrease,onaverage,by24.975piesperweekforeach$1increaseinsellingprice,netoftheeffectsofchangesduetoadvertisingb2=74.131:saleswillincrease,onaverage,by74.131piesperweekforeach$100increaseinadvertising,netoftheeffectsofchangesduetopricewhere SalesisinnumberofpiesperweekPriceisin$Advertisingisin$100’s.DCOVAUsingTheEquationtoMakePredictionsPredictsalesforaweekinwhichthesellingpriceis$5.50andadvertisingis$350:Predictedsalesis428.62piesNotethatAdvertisingisin$100s,so$350meansthatX2=3.5DCOVAPredictionsinExcelusingPHStatPHStat|regression|multipleregression…Checkthe“confidenceandpredictionintervalestimates”boxDCOVAInputvaluesPredictionsinPHStat(continued)
PredictedYvalue<ConfidenceintervalforthemeanvalueofY,giventheseXvaluesPredictionintervalforanindividualYvalue,giventheseXvaluesDCOVAPredictionsinMinitabInputvaluesPredictedValuesforNewObservations
NewObs
Fit
SEFit
95%CI
95%PI
1
428.6
17.2
(391.1,466.1)
(318.6,538.6)
ValuesofPredictorsforNewObservations
NewObs
Price
Advertising
1
5.50
3.50ConfidenceintervalforthemeanvalueofY,giventheseXvaluesPredictionintervalforanindividualYvalue,giventheseXvaluesDCOVATheCoefficientofMultipleDetermination,r2ReportstheproportionoftotalvariationinYexplainedbyallXvariablestakentogetherDCOVARegressionStatisticsMultipleR0.72213RSquare0.52148AdjustedRSquare0.44172StandardError47.46341Observations15ANOVA
dfSSMSFSignificanceFRegression229460.02714730.0136.538610.01201Residual1227033.3062252.776Total1456493.333
CoefficientsStandardErrortStatP-valueLower95%Upper95%Intercept306.52619114.253892.682850.0199357.58835555.46404Price-24.9750910.83213-2.305650.03979-48.57626-1.37392Advertising74.1309625.967322.854780.0144917.55303130.7088852.1%ofthevariationinpiesalesisexplainedbythevariationinpriceandadvertisingMultipleCoefficientof
DeterminationInExcelDCOVAMultipleCoefficientof
DeterminationInMinitabTheregressionequationisSales=307-25.0Price+74.1Advertising
Predictor
Coef
SECoef
T
PConstant 306.50
114.302.68
0.020Price -24.98
10.83
-2.31
0.040Advertising 74.13
25.97
2.85
0.014
S=47.4634
R-Sq=52.1%
R-Sq(adj)=44.2%
AnalysisofVariance
Source
DF
SS
MS
F
PRegression
2
29460
14730
6.54
0.012ResidualError
12
27033
2253Total
14
5649352.1%ofthevariationinpiesalesisexplainedbythevariationinpriceandadvertisingDCOVAAdjustedr2r2neverdecreaseswhenanewXvariableisaddedtothemodelThiscanbeadisadvantagewhencomparingmodelsWhatistheneteffectofaddinganewvariable?WeloseadegreeoffreedomwhenanewXvariableisaddedDidthenewXvariableaddenoughexplanatorypowertooffsetthelossofonedegreeoffreedom?DCOVAShowstheproportionofvariationinYexplainedbyallXvariablesadjustedforthenumberofX
variablesused
(wheren=samplesize,k=numberofindependentvariables)PenalizesexcessiveuseofunimportantindependentvariablesSmallerthanr2UsefulincomparingamongmodelsAdjustedr2(continued)DCOVARegressionStatisticsMultipleR0.72213RSquare0.52148AdjustedRSquare0.44172StandardError47.46341Observations15ANOVA
dfSSMSFSignificanceFRegression229460.02714730.0136.538610.01201Residual1227033.3062252.776Total1456493.333
CoefficientsStandardErrortStatP-valueLower95%Upper95%Intercept306.52619114.253892.682850.0199357.58835555.46404Price-24.9750910.83213-2.305650.03979-48.57626-1.37392Advertising74.1309625.967322.854780.0144917.55303130.7088844.2%ofthevariationinpiesalesisexplainedbythevariationinpriceandadvertising,takingintoaccountthesamplesizeandnumberofindependentvariablesAdjustedr2inExcelDCOVAAdjustedr2inMinitabTheregressionequationisSales=307-25.0Price+74.1Advertising
Predictor
Coef
SECoef
T
PConstant 306.50
114.302.68
0.020Price -24.98
10.83
-2.31
0.040Advertising 74.13
25.97
2.85
0.014
S=47.4634
R-Sq=52.1%
R-Sq(adj)=44.2%
AnalysisofVariance
Source
DF
SS
MS
F
PRegression
2
29460
14730
6.54
0.012ResidualError
12
27033
2253Total
14
5649344.2%ofthevariationinpiesalesisexplainedbythevariationinpriceandadvertising,takingintoaccountthesamplesizeandnumberofindependentvariablesDCOVAFTestforOverallSignificanceoftheModelShowsifthereisalinearrelationshipbetweenalloftheXvariablesconsideredtogetherandYUseF-teststatisticHypotheses:H0:β1=β2=…=βk=0(nolinearrelationship)H1:atleastoneβi≠0(atleastoneindependent variableaffectsY)
IstheModelSignificant?DCOVAFTestforOverallSignificanceTeststatistic:
whereFSTAThasnumeratord.f.=kand denominatord.f.=(n–k-1)
DCOVARegressionStatisticsMultipleR0.72213RSquare0.52148AdjustedRSquare0.44172StandardError47.46341Observations15ANOVA
dfSSMSFSignificanceFRegression229460.02714730.0136.538610.01201Residual1227033.3062252.776Total1456493.333
CoefficientsStandardErrortStatP-valueLower95%Upper95%Intercept306.52619114.253892.682850.0199357.58835555.46404Price-24.9750910.83213-2.305650.03979-48.57626-1.37392Advertising74.1309625.967322.854780.0144917.55303130.70888(continued)FTestforOverallSignificanceInExcelWith2and12degreesoffreedomP-valuefortheFTestDCOVAFTestforOverallSignificanceInMinitabTheregressionequationisSales=307-25.0Price+74.1Advertising
Predictor
Coef
SECoef
T
PConstant 306.50
114.302.68
0.020Price -24.98
10.83
-2.31
0.040Advertising 74.13
25.97
2.85
0.014
S=47.4634
R-Sq=52.1%
R-Sq(adj)=44.2%
AnalysisofVariance
Source
DF
SS
MS
F
PRegression
2
29460
14730
6.54
0.012ResidualError
12
27033
2253Total
14
56493With2and12degreesoffreedomP-valuefortheFTestDCOVAH0:β1=β2=0H1:β1andβ2notbothzero
=.05df1=2df2=12TestStatistic:Decision:Conclusion:SinceFSTATteststatisticisintherejectionregion(p-value<.05),rejectH0ThereisevidencethatatleastoneindependentvariableaffectsY0
=.05F0.05=3.885RejectH0DonotrejectH0CriticalValue:F0.05=3.885FTestforOverallSignificance(continued)FDCOVATwovariablemodelYX1X2YiYi<x2ix1iThebestfitequationisfoundbyminimizingthesumofsquarederrors,e2SampleobservationResidualsinMultipleRegressionResidual=ei=(Yi–Yi)<DCOVAMultipleRegressionAssumptionsAssumptions:TheerrorsarenormallydistributedErrorshaveaconstantvarianceThemodelerrorsareindependentei=(Yi–Yi)<Errors(residuals)fromtheregressionmodel:DCOVAResidualPlotsUsed
inMultipleRegressionTheseresidualplotsareusedinmultipleregression:Residualsvs.YiResidualsvs.X1iResidualsvs.X2iResidualsvs.time(iftimeseriesdata)<UsetheresidualplotstocheckforviolationsofregressionassumptionsDCOVAUsettestsofindividualvariableslopesShowsifthereisalinearrelationshipbetweenthevariableXjandYholdingconstanttheeffectsofotherXvariablesHypotheses:H0:βj=0(nolinearrelationship)H1:βj≠0(linearrelationshipdoesexist betweenXjandY)AreIndividualVariablesSignificant?DCOVAH0:βj=0(nolinearrelationshipbetweenXjandY)H1:βj≠0(linearrelationshipdoesexist betweenXjandY)TestStatistic: (df=n–k–1)AreIndividualVariablesSignificant?(continued)DCOVARegressionStatisticsMultipleR0.72213RSquare0.52148AdjustedRSquare0.44172StandardError47.46341Observations15ANOVA
dfSSMSFSignificanceFRegression229460.02714730.0136.538610.01201Residual1227033.3062252.776Total1456493.333
CoefficientsStandardErrortStatP-valueLower95%Upper95%Intercept306.52619114.253892.682850.0199357.58835555.46404Price-24.9750910.83213-2.305650.03979-48.57626-1.37392Advertising74.1309625.967322.854780.0144917.55303130.70888tStatforPriceistSTAT=-2.306,withp-value.0398tStatforAdvertisingistSTAT=2.855,withp-value.0145(continued)AreIndividualVariablesSignificant?ExcelOutputDCOVAAreIndividualVariablesSignificant?MinitabOutputTheregressionequationisSales=307-25.0Price+74.1Advertising
Predictor
Coef
SECoef
T
PConstant 306.50
114.302.68
0.020Price -24.98
10.83
-2.31
0.040Advertising 74.13
25.97
2.85
0.014
S=47.4634
R-Sq=52.1%
R-Sq(adj)=44.2%
AnalysisofVariance
Source
DF
SS
MS
F
PRegression
2
29460
14730
6.54
0.012ResidualError
12
27033
2253Total
14
56493tStatforPriceistSTAT=-2.31,withp-value.040tStatforAdvertisingistSTAT
=2.85,withp-value.014DCOVAd.f.=15-2-1=12=.05t/2=2.1788InferencesabouttheSlope:
t
TestExampleH0:βj=0H1:βj
0Theteststatisticforeachvariablefallsintherejectionregion(p-values<.05)ThereisevidencethatbothPriceandAdvertisingaffectpiesalesat
=.05FromtheExceloutput:RejectH0foreachvariableDecision:Conclusion:RejectH0RejectH0a/2=.025-tα/2DonotrejectH00tα/2a/2=.025-2.17882.1788ForPricetSTAT=-2.306,withp-value.0398ForAdvertisingtSTAT
=2.855,withp-value.0145DCOVAConfidenceIntervalEstimate
fortheSlopeConfidenceintervalforthepopulationslopeβjExample:Forma95%confidenceintervalfortheeffectofchangesinprice(X1)onpiesales:-24.975±(2.1788)(10.832)Sotheintervalis(-48.576,-1.374)(Thisintervaldoesnotcontainzero,sopricehasasignificanteffectonsales)
CoefficientsStandardErrorIntercept306.52619114.25389Price-24.9750910.83213Advertising74.1309625.96732wherethas(n–k–1)d.f.Here,thas(15–2–1)=12d.f.DCOVAConfidenceIntervalEstimate
fortheSlopeConfidenceintervalforthepopulationslopeβjExample:Exceloutputalsoreportstheseintervalendpoints:Weeklysalesareestimatedtobereducedbybetween1.37to48.58piesforeachincreaseof$1inthesellingprice,holdingtheeffectofadvertisingconstant
CoefficientsStandardError…Lower95%Upper95%Intercept306.52619114.25389…57.58835555.46404Pric
温馨提示
- 1. 本站所有资源如无特殊说明,都需要本地电脑安装OFFICE2007和PDF阅读器。图纸软件为CAD,CAXA,PROE,UG,SolidWorks等.压缩文件请下载最新的WinRAR软件解压。
- 2. 本站的文档不包含任何第三方提供的附件图纸等,如果需要附件,请联系上传者。文件的所有权益归上传用户所有。
- 3. 本站RAR压缩包中若带图纸,网页内容里面会有图纸预览,若没有图纸预览就没有图纸。
- 4. 未经权益所有人同意不得将文件中的内容挪作商业或盈利用途。
- 5. 人人文库网仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对用户上传分享的文档内容本身不做任何修改或编辑,并不能对任何下载内容负责。
- 6. 下载文件中如有侵权或不适当内容,请与我们联系,我们立即纠正。
- 7. 本站不保证下载资源的准确性、安全性和完整性, 同时也不承担用户因使用这些下载资源对自己和他人造成任何形式的伤害或损失。
最新文档
- UNIT2 单元检测(二)2024-2025学年高中英语选择性必修第一册同步导学案配套教学设计(人教版2019)
- 2022年广东省中考满分作文《我不由得加快了脚步》2
- 五年级下册 第二十七课 烦恼来了怎么办 教案
- 名校联盟陕西省山阳县城关镇第一初级中学初中体育教学设计(3份)
- 4 团团圆圆过中秋 第二课时 教学设计-2024-2025学年道德与法治二年级上册统编版
- 地球的圈层结构教学设计 中图版
- Unit 5 What color is it Lesson 27 (教学设计)-2023-2024学年人教精通版英语三年级上册
- 小学生心理健康促进活动安排
- 卫风·氓教案 苏教版
- 八年级语文下册 第一单元 第1课《社戏》同步教案 新人教版
- 2024-2030年海参行业市场深度调研及发展规划与投资前景研究报告
- 2024义务教育数学新课标课程标准2022版考试真题附答案
- 二型呼吸衰竭的课件
- 2022年版初中化学课程标准新课标考试题库及答案1
- 螺杆空压机微电脑控制器MAM-KY02S(B)-(Ⅷ)型用户手册
- 2024中华人民共和国农村集体经济组织法详细解读课件
- 2024年中国兵器工业集团限公司夏季招聘(高频重点提升专题训练)共500题附带答案详解
- 大学劳动教育(高等院校劳动教育课程)全套教学课件
- 《幼儿园中班第一学期家长会》 PPT课件
- 一元一次方程评课材料
- B4 河南省高中毕业生登记表
评论
0/150
提交评论