View Full Version : how do you test for heteroscedasticity in a multiple regression analysis?
03-14-2006, 08:27 PM
my class didnt go into much specifics on this, but basically i ran a regression in Excel with one dependent var and 4 Independet vars...how do you test for heteroscedasticity in a multiple regression analysis given the output from excel?
03-15-2006, 07:16 AM
03-15-2006, 01:22 PM
Don't you just do it like you normally would? Plot the standardized residuals vs predicted "y". Maybe this will help you: http://pareonline.net/getvn.asp?v=8&n=2 , I just got it after a quick google search. Let me know if this is not what you meant
03-15-2006, 03:01 PM
There is a complete chapter referenceing this in Pindyck and Rubinfeld. Chapter 6.1.2 152-159.
vBulletin® v3.7.6, Copyright ©2000-2013, Jelsoft Enterprises Ltd.