53%
43%. Don’t see it being higher then L1, those folks got a gift compared to those of us that took and passed (or failed) in June/Dec 08!
37%
Just read this blog entry on Schweser regarding test results for L1: http://www.schweser.com/cfa/blog/index.php?item_id=707&new_level=cfa1&cfa_blog=cfa1 Looks like Schweser believes the upward trend for L1 result is due to the change from 4 stems to 3 stems in the MC. (The dropping enrollment could be a cause too.) So I can only speculate the pass rate for L2 this year could follow a similar trend (it would be upward biased). I am guessing somewhere between 50-55% this year.
I did a little regression analysis on all the post-millennium LI vs. LII scores and found this as my output: y = 0.4066x + 0.2951 Therefore, if we plug the 46% LI got for June, we get a predicted result for LII 2009 of… 48%!!! The R2 = 0.1263, so this could come in way above or below, but I think this will probably be the first and last time I use any of that quant BS!
You probably should use a lagged L1, not same year. In fact, to get re-takers, plus first time takers, you’d have to consider both L1 and L2 lagged rates.
What dlpicket said
I’ll say 57%. I give myself 50/50 odds. I felt that way as soon as I walked out of the test, and after the months of going through the spreadsheets and pouring over the questions again, I still don’t have the slightest idea. 50/50, a flip of a coin. This day can’t end soon enough.
I think pass rate data have seasonality in it. (When the investment industry is bullish, you expect the pass rate to drop and vice versa.) So to properly model the time series, you would need to introduce lag terms. Note: be careful not to sample data when there is a structural change in the data. You may need more than one time series. I’m not too sure how to model the retakers. Intuitively, the percentage of enrolled retakers could introduce heteroscedasticity and you need to clean up the data first before modeling the series.
dlpicket Wrote: ------------------------------------------------------- > You probably should use a lagged L1, not same > year. In fact, to get re-takers, plus first time > takers, you’d have to consider both L1 and L2 > lagged rates. I was thinking the same thing.
JensensalphaMale Wrote: ------------------------------------------------------- > dlpicket Wrote: > -------------------------------------------------- > ----- > > You probably should use a lagged L1, not same > > year. In fact, to get re-takers, plus first > time > > takers, you’d have to consider both L1 and L2 > > lagged rates. > > > I was thinking the same thing. but you also have to remember there could’ve been more ppl studying for this exam who were also laid off earlier in the year.
0 %
Ok. Ran 1995 to 2005 with lagged L1 and L2. I select 95 as a starting point because that’s when they started doing Angoff grading. here is our regression formula L2(1) = 30.25 + L1 (0) * 0.48 - L2(0) * 0.05 I wouldn’t read too much into it. There is a LOT of standard error in there.
I am predicting level 2 pass rate will be 55 percent and level 3 going to be at 75 percent. This is my upward revision from historical pass rates of 45 and 61 respectively. Level pass rate jumped from 35 percent last year or 37.5 Percent from last 5 years average to 46 percent. If you extrapolate it to level 2 and 3 it should be 55 and 75 percent. However level 3 has descriptive part in it, so let us lower level 3 to 69 percent.
I don’t think simple extrapolation from L1 onto L2 and L3 would work. This is because the exam formats are very different. You could still model the same type of impact on L2 and L3 (in this case, the changes from 4 stems to 3 stems). But I would model all three levels as three different time series.
Christmas is arriving early this year, 100% pass rate. Santa Clause is settling some old grudges against CFA institute.
I did a little regression analysis on all the post-millennium LI vs. LII scores and found this as my output: y = 0.4066x + 0.2951 Therefore, if we plug the 46% LI got for June, we get a predicted result for LII 2009 of… 48%!!! The R2 = 0.1263, so this could come in way above or below, but I think this will probably be the first and last time I use any of that quant BS!
Updated (repeats and ranges are omitted): 0% chiank123 30% Damil4real 35% chowder 37% BiPolarBoyBoston 38% TheAliMan 39% evgong 40% sublimity 41% wake2000 42% nirjraina 43% JensensalphaMale 44% mdecav 45% adalfu 46% london1 47% pjc2000 48% agrover 49% Alchemist1320 50% akinzo 51% bos_IT_guy 52% CF_AHHHHHHHHH 53% ILFXAM 54% 4Tay 55% bid_offer_shark 56% gdoggwins 57% SmokeyJoeWood 58% DanLieb 60% oldboy321 89% golfer 99% charlottekid 100% iblees82
If I fail I really hope the passing rate is under 45 percent haha
why does everyone want it to be so low?