When utilizing a proxy for one or more independent variables in a multiple regression model, which of the following errors is most likely to occur? A) Heteroskedasticity. B) Multicollinearity. C) Serial correlation. D) Model misspecification.
B or D. will go with B.
I’d go with D as long as you are not also putting in the proxied independent variable in the model in which case you would likely get some serious B.
D. This is an example of using an independent variable that is (potentially) measured with error, which is a model misspecification. An example of this is using a “free float” variable which is a proxy for corp. governance quality. Using such a proxy can be erroneous.
B but I’m not sure if cfasf1 convinced me or that I thought of it myself…
D, what if your chosen proxy sucks at being a proxy.
It’s D for sure. Proxies of independent variable will always cause Model mispectification. Ya!! sneaked in from work to answer this one 
mcpass, this is what happens when you let me convince you into anything. lol.
All right! D is the correct answer. Zombie, thanks for that explanation. I wanted to re-confirm my suspicions.
I’m glad you’re not sitting close to me in June 
Haha. Continents apart will keep you safe mc.
i’ll admit. i have no idea even though i’ve done quant a couple of times. i think it’s the wording. what does the question mean? is it if you have two independent variables basically measuring the same thing? or is it omitting an important indy variable?.. i think i understand the concepts, but i don’t understand what they mean by their wording thanks in advance for any help!!! and of course, it’s very likely i don’t understand the basic concept.
I’m definitely going with D) I remember this is straight out of the readings. * Sweet, one quant question that I would actually get