I wanted to start a discussion on Nassim Taleb's Empirica (his hedge fund) and, more importantly, his trading strategy. My understanding is that he has taken the opposite approach of Victor Niederhoffer. That is, instead of selling OTM premium, he is buying it. It seems that he believes that most traders underestimate the probability of a large adverse move and, in the long run, it is better to be "long the tails", rather than short. Empirica's strategy seems to be contrarian in nature and that's one reason that I find it to be interesting (though I'm not sure if it can be profitable over the long run). It seems that Empirica is willing to take many small loses while waiting for a big move, a "home run", that will not only cover their previous losses but provide additional gains as well. If this is the case, Empirica will never "blow up" in the sense that one drastic event won't take them out (unless I'm missing something). The worse thing that could happen is that they will simply "bleed to death" over a long period of time. So, a few questions for the group... 1) Do you think Nassim Taleb is right in that traders underestimate the probability of a drastic move in the underlying product? 2) Can a strategy like this work over the long run? 3) If you had to choose one, would you rather be short the tails (Niederhoffer) or long the tails (Taleb)? -SG
Sleeping Giant: Statistically, Taleb will prove to be correct IF he (and his staff) can be disciplined enough to only take positions where his local losses can be "subsidized" by inflows or his working account's risk free income. Anyone can do this, especially when volatility is low as it is today. Simply monitor Implied Vol, looking for instruments that have unusually low absolute volatility, and buy treasuries to pay for the "rent". Sooner or later "boom" a six sigma event will ring your register. Best Regards, Steve46
excellent thread-- attached please find a background article for those of you not familiar with the names mentioned : http://www.gladwell.com/2002/2002_04_29_a_blowingup.htm enjoy !!!!! surfer
There has got to be a middle ground between the polar opposites of Neidehoffer and Taleb's strategy. If I had to pick a side, I would side with Taleb since you can "amortize" our losses which can be paid for by gamma scalps. Trick is to discern when to put on the tail position. One of the traders in the stock maket wizards had a great discourse on this . This book by J schwager is 80% fluff save for a couple of traders inclu option trader who structures his spreads basis tech analysis.
What little research I have done has convinced me that buying options in general tends to be a losing proposition. To profit, it seems that you have to be extremely discerning. I think the art to Taleb's strategy is to monitor multiple underlyings and wait for a large enough safety buffer of mispricing to come along before buying a tail. I do think that implied vols have little predictive power with regards to multi-sigma events. ie multi-sigma events occur with the same distribution regardless of whether you are in a low vol or high vol environment, thus the expected value of buying tails goes up dramatically in a low vol environment. This is merely a supposition without any proof.
Strangely enough I just read this article earlier today. My only question was that considering it requires consistent losses for a long period of time until a large profitable trade, how is the hedge fund able to retain investors/capitol. Wouldn't most investors simply see continual losses and decide to get out of the fund. Clearly any investor should know the philosophy of the fund they are investing in, but like the article said, it is very hard to be the person who slowly bleeds. How many times can a Sept.11 happen? After it does happen, what is the chance that it will happen again within a brief enough time period that your capitol will not be reduced to zero? Considering the firm is taking losses everyday, once they have a big win, the win will be from a smaller position size than originally possible. Isn't there a chance that the win won't compensate for those losses? -ae_trading
I believe this strategy will make money over time, however the rate of returns are not that great. This strategy makes sense for large Hedge funds which can live off the 1-2% management fee, but for individual traders very hard to make ends meet. Empirica's performance numbers are not published, just wondering if anyone knows how they have done over the years?
Taleb's strategy is very similar to classis trend following strategy used by most commodity fund managers for example the turtle system, lots of small losers but then some bets will reap huge returns. It does work ! Taleb is something of a celebrity in the Hedge fund speaking circuit, his book is a big hit, gives him a lot of credibility. I am not sure what his performance is like, but there is a market for his product. Some investors are willing to wait for small consistent returns.
IMO the key point about fat tails is not that they are perpetually underpriced, but that occasionally they are underpriced by a huge amount. The obvious "holy grail" of fat tails is to be able to identify when the probability of them occuring is much higher (or lower) than usual. If you have some method which can identify potential crises ahead of the fact, and/or to avoid potential stable low volatility periods, then you have a massive edge. I certainly think it's possible to develop such a method.