资源描述：
《多元线性回归模型估计》由会员上传分享，免费在线阅读，更多相关内容在教育资源-天天文库。
1、MultipleRegressionAnalysisy=b0+b1x1+b2x2+...bkxk+u1.Estimation1ParallelswithSimpleRegressionb0isstilltheinterceptb1tobkallcalledslopeparametersuisstilltheerrorterm(ordisturbance)Stillneedtomakeazeroconditionalmeanassumption,sonowassumethatE(u
2、x1,x2,…,xk)=0Stillminimizingthesumofsquaredresiduals,s
3、ohavek+1firstorderconditions2theOLSregressionlineorthesampleregressionfunction(SRF).istheOLSinterceptestimateandaretheOLSslopeestimates.Stilluseordinaryleastsquarestogettheestimates:3OLSFirstOrderConditionsThisminimizationproblemcanbesolvedusingmultivariablecalculas.Thisleadstok+1linearequationin
4、k+1unknown:……4AFittedorPredictedValueForobservationi,thefittedvalueisTheresidualforobservationiisdefinedasinthesimpleregressioncase,Theproperties123Thepoint()isalwaysontheOLSregressionline:5InterpretingMultipleRegression6A“PartiallingOut”Interpretation7“PartiallingOut”continuedPreviousequationimp
5、liesthatregressingyonx1andx2givessameeffectofx1asregressingyonresidualsfromaregressionofx1onx2Thismeansonlythepartofxi1thatisuncorrelatedwithxi2arebeingrelatedtoyisowe’reestimatingtheeffectofx1onyafterx2hasbeen“partialledout”8SimplevsMultipleRegEstimate9Goodness-of-Fit10Goodness-of-Fit(continued)
6、Howdowethinkabouthowwelloursampleregressionlinefitsoursampledata?Cancomputethefractionofthetotalsumofsquares(SST)thatisexplainedbythemodel,callthistheR-squaredofregressionR2=SSE/SST=1–SSR/SST11Goodness-of-Fit(continued)12MoreaboutR-squaredR2canneverdecreasewhenanotherindependentvariableisaddedtoa
7、regression,andusuallywillincreaseBecauseR2willusuallyincreasewiththenumberofindependentvariables,itisnotagoodwaytocomparemodels13AssumptionsforUnbiasednessPopulationmodelislinearinparameters:y=b0+b1x1+b2x2+…+bkxk+uWecanusearandomsampleofsizen,{(xi1,xi2,…,xik,yi):i=1,2,…,n},fromthepopulationmodel,
8、sothatthesamplemodelisyi=b0+b1xi1+b2xi2+…+bkxik+uiE(u
9、x1,x2,…xk)=0,implyingthatalloftheexplanatoryvariablesareexogenousNoneofthex’sisconstant,andtherearenoexactlinearrelationshipsamongthem14TooManyorTooFewVariablesWhat
