max drawdown

CFA Level III
#21

i was just going to say we should get JDV in on this. i guess bigwilly is right…

#22

i asked this question a while ago and seemed to get the consensus that it was (in your example) 8. I was also screwed up by the Schweser chart. The book/Schweser seem to indicate over and over that it is the HWM to low point until next HWM, so it fits w/ BW example above.

#23

Awesome, I’m on Google! Cville I think your IR person was exhibiting, Illusion of Knowledge b/c she has her charter…haha only joking.

#24

Just to clarify: for the numbers given above do you guys agree on 1. 80% (from 10 to 2) 2. 50% (8 to 4 or 4 to 2) ? Thanks

#25

The Maximum Drawdown Period is the worst period of “peak to valley” performance for the series, regardless of whether or not the drawdown consisted of consecutive months of negative performance, or where the Max Period falls within the range of data.

#26

i think the problem here is the schweser graph: check the text and the formula, they do MONTHLY average return divided by MONTHLY max draw down i guess that the “tenor” of the max draw down must be equal to the one you use in the numerator. if we use yearly return, yearly max draw down. if we use average monthly return, monthly drawdown… apples to apples in an extreme, the longer the observation period, the higher the max drawdown can be. it makes no sense to compare a higher and higher max drawdown with a monthly (or static tenor) average return

#27

hala_madrid Wrote: ------------------------------------------------------- > i think the problem here is the schweser graph: > check the text and the formula, they do MONTHLY > average return divided by MONTHLY max draw down > > i guess that the “tenor” of the max draw down must > be equal to the one you use in the numerator. if > we use yearly return, yearly max draw down. if we > use average monthly return, monthly drawdown… > apples to apples > > in an extreme, the longer the observation period, > the higher the max drawdown can be. it makes no > sense to compare a higher and higher max drawdown > with a monthly (or static tenor) average return I agree with this. Sounds like a great answer to me. Thanks!

#28

IMHO MDD is a max delta between t0 and t1. So max is 4, in this scenario

#29

i’m going to concede – admit a devastating defeat; i think in my IR persons defense, i probably phrased the question the way i wanted it answered - - my FoF buddy threatened to pull his money 1) bc i’m an idiot and 2) bc i told him i was on a blog where we needed the info

#30

haha

#31

i hope you didn’t specifically say “yeah this guy bigwilly says…”

#32

I thought you might find this useful. This is actual mutual fund data over the last year. I ran this on some software at work to help you guys out. Mutual Fund Return, % Total 7-Jul -3.2 7-Aug 2.59 Max Drawdown period Nov-07 - Mar-08 7-Sep 4.17 Max Drawdown Return -21.78 7-Oct 6.47 Max Drawdown Duration 5 7-Nov -5.64 7-Dec -0.28 8-Jan -10.28 8-Feb -5.24 8-Mar -2.23 8-Apr 4.81

#33

so what would happen if May 8 data was down 10%; would that mean max drawdown would included that data or does your program reset at april - - your data has the data as a continuous stream of losses which is the bit in question - - i’m guessing duration changes to 7 despite the April up month?