Your point 3 is very difficult to quantify, because it is very idiosyncratic. The prepayment can either be due to market factors like interest rates or life events such as job relocation, I am not even sure if banks record this kind of data.
I decided to take a cycle off to play golf and enjoy the beach, but plenty of people have done May and Nov back to back. There is still time now, but you should really decide within the next week or so.
Interesting, I did not read the question the same way. My thought process was that the depositor may have been lured by higher rates from competitors, but chose to stay at the same bank, so these deposits were pretty "sticky." Hence, when rates rose these deposits were unaffected. Given the...
50% is roughly the historical pass rate. It doesn't have to be that, could be higher or lower. What I'm trying to say is, if you want to go with a historical simulation method, then 50% is your starting point which is what I assumed the other fellow did.
This differs from your assumption of...
I think you fell for the same trap I did. The regression coefficient was -.37 something, and is the negative of mean reversion rate. So the autocorrelation is .63 or 1-.37. Unless the question was asking for mean reversion rate? I really don't remember at this point
Well I think 40 is 50% correct, yes? Based on passage rates of around 50%, and assuming a normal distribution of scores, the cutoff should be about half. But then again, we do not know the underlying distribution, maybe it is highly skewed to the right? Who knows