Page 33 - Regression Guideline

Important"Terms"You"Need"to"Know"
In"this"hypotheJcal"scenario,"the"nonPindependence"of"bedrooms"and"gross"living"area"must" be"taken"into"account"when"considering"how"this"influences"price"as"shown"in"the"diagram"to" the"right."A"mulJvariable"(but"not"bivariate)"regression"model"will"take"into"account"the"nonP independence"of"(higher"order"correlaJons"in"staJsJcal"parlance)"the"property"characterisJcs" when"esJmaJng"the"appraised"value"of"a"property.""
&
Stepwise&Regression.&An"important"part"of"regression"modeling"is"determining"which"set"of" predictors"do"the"best"job"of"predicJng"the"dependent"variable."SAVVI"AnalyJcs"uses"a"method" called"stepwise"regression."This"method"examines"each"individual"property"characterisJc"as"a" predictor"to"determine"the"extent"to"which"knowing"this"specific"characterisJc"was"useful"for" gecng"a"more"accurate"esJmate"of"the"sales"price"for"the"properJes"examined.""
"
The"“first"step”"in"a"stepwise"regression"idenJfies"the"“best”"onePPPvariable"model."That"is," what"single"property"characterisJc"best"predicts"sales"price"from"among"the"characterisJcs" being"considered."Subsequent"steps"will"idenJfy"the"“best”"twoPPP"variable,"threePPPvariable," etc."models"unJl"the"enJre"set"of"predictors"is"considered."Only"those"predictors"that"improve" the"accuracy"of"the"appraised"price"are"included"at"each"subsequent"step.""
"
To"determine"the"bestPficng"egression"model"(i.e.,"how"well"it"“fits”"the"data)"a"number"of" staJsJcs"are"used"to"compare"models"with"different"subsets"of"predictors."The"“best”"models" are"idenJfied"as"those"that"contain"the"subset"of"predictors"that"are"staJsJcally"reliable"(the"FP test"of"staJsJcal"significance),"obtains"the"maximum"R2"(the"amount"of"variance"in"sales"price" explained),"and"is"accurate"(using"another"staJsJcal"test"called"“Mallow’s"Cp”").""
"
SelecJon"of"variables"in"a"stepwise"regression"can"be"either"forward"or"backward."In"forward" selecJon,"the"model"starts"with"zero"predictors"and"adds"one"new"predictor"at"a"Jme." Variables"conJnue"to"be"added"to"the"model"unJl"no"remaining"variables"improves"the"value" of"R2"beyond"a"minimum"threshold"of"staJsJcal"significance."(The"default"minimum"“FPPPtoPPP enter”"in"SAVVI"is"0.15.)"Backward"eliminaJon"is"similar"except"that"the"model"begins"with"all" predictors"included"and"then"eliminates"those"that"do"not"maintain"a"staJsJcal"threshold"of" significance.""
"
32"