Chapter 1 Matrix approach to Simple Regression Models

Chapter 1 Matrix approach to Simple Regression Models -...

Info icon This preview shows pages 1–15. Sign up to view the full content.

CYM 1.1 Chapter 1 MatrixApproachto SimpleRegressionModel ST3131 Regression Analysis
Image of page 1

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

CYM 1.2 Overview Leastsquaresestimation Sumofsquares:SST,SSR,andSSE ANOVAtable, F ‐test Varianceof Confidenceintervalsfor β 0 and β 1 Confidenceintervalfor ݕ | ݔ Predictionintervalfor ST3131 Regression Analysis
Image of page 2
CYM 1.3 1.1 Simple Regression Model Considerthesimpleregressionmodel where ݅ istheresponse(ordependent)variable, ݅ isthepredictor(orindependent)variable, and areunknownparameters, sareindependentnormalrandomvariables, with and forall i . ST3131 Regression Analysis Deterministic but unknown Random
Image of page 3

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

CYM 1.4 Simple Regression Model ST3131 Regression Analysis Random Error of Y for this X i value Y X Observed Value of Y for X i Expected Value of Y for X i 0 1 Y β β X X i Slope = β 1 Intercept = β 0 ε i i 0 1 i i Y β β X ε
Image of page 4
Simple Regression Model Recall TheLeastSquaresEstimatesfor and inModel(1)are obtainedby ,ఉ TheLSEfor and aregivenby ௜ୀଵ ௜ୀଵ CYM 1.5 ST3131 Regression Analysis
Image of page 5

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

Simple Regression Model Recall WewanttotestH 0 : againstH 1 : Teststatistic ି଴ ௏௔௥ RejectH 0 if where with CYM 1.6 ST3131 Regression Analysis
Image of page 6
Simple Regression Model Model(1)canbeexpressedinamatrixformasfollows: where , , ,and with and . Hence and . CYM 1.7 ST3131 Regression Analysis
Image of page 7

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

1.2 Least Squares Estimation Leastsquaresestimatorfor β canbeobtainedby minimizing (i.e. ௜ୀଵ ). Since ,therefore ST3131 Regression Analysis CYM 1.8
Image of page 8
Least Squares Estimation (Continued) Rewritetheaboveexpressionas Byexpandingtheaboveexpression,wehave Hence and ST3131 Regression Analysis CYM 1.9
Image of page 9

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

Least Squares Estimation (Continued) Assumingthat isnon‐singular(i.e. ିଵ exists), then ିଵ and ST3131 Regression Analysis CYM 1.10
Image of page 10