Monki Ma, CFA lvl 2 candidate is performing a Durbin-Watson Test for a regression model for his manager. The DW result is in between the low critical value and high critical value, which is inconclusive, what should Monki do? A) Use Handson method to adjust standard errors B) Conclude that it has problem with serial-correlation since DW value is not in the reject zone. C) Ignore serial-correlation as a problem, since DW value is in neither of the positive nor negative autocorrelation zone. D) use another model.
I would have guessed C. But I have a feeling thats a no no in regression analysis.
I’d give A a shot i suppose?
E) Monki should go to bed now, just as planner is…
i go with c
I go with D, if its within that range then the results are inconclusive. I don’t think inconclusive would mean to ignore it.
BEFORE WE GO TOO DEEP INTO THIS… i want to say that I don’t have an answer… I am really looking for help on this question! AND I MADE THAT QUESTION UP, so perhaps all 4 answers are wrong lol
CFAI and schweser say nothing, at least that my tired mind can find. But… Googled it and found along study. Page 16 has a little on it. http://www.seas.upenn.edu/~ese302/Projects/Project_3.pdf