LOS 12 G/H Heteroskedasticity/Multicollinearity

Hi all, LOS G–Discuss the types of heteroskedasticity and the effects of heteroskedasticity and serial correlation on statistical inference. LOS H–Describe multicollinearity and discuss its causes and effects iin regression analysis. Both Schwesser and the CFAI texts seem to go beyond the LOS and include a section on “correcting heteroskedasticity/serial correlation/multicollinearity.” Obviously, i’m going to study the “correcting” section like it’s fair game, but did anyone else find this funny? I know it can be dangerous to take the LOS’s too literally, but when I was reading about Hansen’s method etc I initially felt like i didn’t need to know it because it wasn’t in the LOS. Shouldn’t the LOS also say “Describe methods of correcting these violations” instead of simply “discuss the types”

If I remember correctly, “correcting” was a LOS in the 2007 exam…

Thanks, Anyone else? I type out heteroskedasticity like 4 times :slight_smile:

jabroni, darkstar is correct. Here’s a link to our recent discussion of this question. http://www.analystforum.com/phorums/read.php?12,647119

Do we have to know the formulas to test for heteroskedasticity or the DW formula for Serial correlation?

No. Nobody knows those formulas except computers (which is what they are for). Know what these things do to your regression rsults including your model parameters and overall goodness of fit measures. Know what to do about it if you have it.