Take the 2-minute tour ×
Quantitative Finance Stack Exchange is a question and answer site for finance professionals and academics. It's 100% free, no registration required.

How would one go about cloning/replicating returns of a hedge fund or a strategy. That is given a return series of the object to be clone, is it possible to decompose return and reconstruct another different passive strategy that generates return that are correlated above 75%+? Perfect replication is definitely not possible, but I wanted to understand the process and possibly the success people have had in this area.

My initial thought on such subject would be to decompose its return using factor models...

Some pictures (source: bwater): Edit: 2012-12-14enter image description here enter image description here

enter image description here

enter image description here

share|improve this question
    
Reliable 75%+ correlation is definitely not possible for the vast majority of hedge fund types. –  Jase Dec 13 '12 at 3:26
    
i beg to differ, there are lot of hedge funds that are dominated by beta rather than alpha. B-water did two piece of analysis and found that excess hedge fund return is practically flat since '09. Also they were able to replicate some strategy with 80% correlation (ie Emerging market hedge funds) –  user1234440 Dec 13 '12 at 3:54
    
If you're suggesting not being restricted by a style, then yes it would be very easy to weight/select a portfolio to match any number. I'm not sure what forward thinking solutions doing that could possibly bring however. –  jeff m Dec 13 '12 at 4:42
    
I am trying to figure out ways to tap in to different lowly correlated return streams to help diversify existing portfolios. And it seems hedge fund is a good place to start given their diverse market exposure. –  user1234440 Dec 13 '12 at 4:47
1  
Replication via factor or style attribution at rho > 0.8 is not particularly difficult -- over the past. Not so easy looking forward, you'll always be trading last quarter's factor mix. Harry Kat's distributional replication approach might be worth a try, but none of the trading pools/allocations based on it have rung any bells, AFAIK. –  Kevin Schmit Dec 13 '12 at 6:14
show 6 more comments

1 Answer

up vote 3 down vote accepted

You are treading controversial waters. It's hard to summarize, but at the risk of oversimplifying, there are three broad schools of thought:

  1. "Linear Models": Classic Examples are a string of papers from Jasmina Hasanhodzic and Andy Lo at MIT (scholar.google.com should give you plenty). For similar work related to Mutual Funds that you may be able to repurpose you should look at the classic "Sharpe Returns Based Style Analysis (aka RBSA)" upon which most linear approaches are based.
  2. "Non Linear and 'Mystery' models': (i.e. details undisclosed) models. Classic example is from the infinitely entertaining Harry Kat et al (examples include "Tell Me What You Want, What You Really, Really Want!" but if you google it, you'll need to add some sort of hedge fund replication tag to avoid getting nothing but Spice Girls references)
  3. "It's not easy": Classic examples here are hard to find, but the most eloquent (and imho unbiased) are from Amenc et al at EDHEC. Examples include "Performance of Passive Hedge Fund Replication Strategies"

The short version, sadly, is that the general feeling (amongst the majority of academics, at least imho) is that what you are asking about is not easy to truly "clone" :-( but check out the references above anyway, perhaps you'll spot something new and interesting.

Having said that, many academics (and, apparently, some of your commenters) feel that most hedge fund "alpha" is really beta disguised as alpha. In that case, depending on what you mean by "replicating returns of a hedge fund" you may or may not have an plausible task on your hands. Replicating the "beta" portion of a hedge fund may indeed be possible. The classic reference here is probably any of the string of papers by Fung and Hsieh (again, scholar.google.com is your best friend).

Bottom Line: it's a matter of opinion and if you had a more precisely stated question, you might get a more precise answer. I hope that helps at least a little :-)

share|improve this answer
    
thanks for the detailed answer. the reason I think its possible is because hedge fund return is dominated by beta, and therefore it should in theory be replicable to a certain extend. I got the idea when I was reading a book that mentioned standard beta based asset classes' return can be separated between risk free, structural beta, and structural alpha. The only component that offers diversification is the structural alpha, all others that have risk return characters below the cash equity line can be in theory replicated by a mix of cash and equities. I may be interpreting it wrong. –  user1234440 Dec 13 '12 at 15:13
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.