(BQ) Part 2 book "Applied multivariate statistical analysis" has contents: Multivariate linear regression models, principal components, factor analysis and inference for structured covariance matrices; canonical correlation analysis; discrimination and classification; clustering, distance methods, and ordination. | CHAPTER 7 Multivariate Linear Regression Models I NTRODUCTI ON Regrees ion anal(dyependente )stvarta isitaiblcalesmetomhodolcologyel ctfioornpredicting val(uinesdependent) sis is th fr a of of one moriable values. It can also be used for asses ing the effects of the predictor variables varthe responses. Unfortunately, the name culled from the title of the onif rst paper on the subject by F. Galton [14] , in no way reflects either the importance or breIadtthhisofchaptiecrat, weoi nfofirstthdiissmets htodolmultiple regres ion model for the predic appl cu he ogy. tion ofnadependentse . sOur treiats tmhen genertabeil zedsomewhat etetrhsee,prasedia cvasit otnlit r ons model ent mus to handl erseaettuhreefexilo sotwis ong tbooksubj, ienctas. c(endiyoug arorederintoferedistfeidculintpurBowergmreangreandioOn 'analysisl, he s If n suin s Connel [and]5 , Netesrb,erWas[9ns]e,rSeberKut0]n, and, and Nachtgserhei[1m5].[)17]OurDrabbreviandta eSmirt etath m[1ent, Cookh man, [2 er Goldber , y: aper d 2] hig Weithe regrg es ion assumptions and their consequences, alternative formulations lofightthse regres ion model, and the general applicability of regres ion techniques seemingly dif erent situations. or response predictor regression, several single of to TH E CLASSICAL LIN EAR REGRESSION M O D E L • • • LetFor exampl, e,bewithpredic4,towevarmiiagblhtehaveought to be related to a response variable r s th current market value of home Y. z 1 , z2 , Zr r r == Y 354 == Section and The Classica l Linear Reg ress ion M odel 355 sloqcatieonfe(etindiofclativoinrgfoarr ezone of city) uar a zz32 appraised value last year z4 regrqualioitnymodel sttartuects tihoatn (prisiccompossqeuarofeafmean, which depends of cons e per d oot) Thea ccontsal iicnauousnill earmanneres on the z/s, and a random error which accounts for mea isnurement error and the effects of other variables not explicitly considered in the model. Thetivalatuoresarofe .
