財經觀察 2007 --- Eric Rosenfeld Talking LTCM, Ten Years Later

Eric Rosenfeld was a trader and principal in the Long-Term Capital Management hedge fund, a landmark Wall Street disaster.

Prior to LTCM, Rosenfeld was an instructor at Harvard University. About one year after LTCM's rescue, in 1999, he joined John Meriwether as a partner in JWM Partners LLC, which started operations with about $250 million under management. He left JWM Partners to join Paloma Partners, a Greenwich fund-of-funds.

In 2007, Rosenfeld founded Quantitative Alternatives LLC in Rye Brook, NY with Bruce Wilson and Robert Shustak.^[1]

As a graduate student at MIT, he worked with Mitch Kapor, future founder of Lotus, to create and sell a financial statistics program written in BASIC for an Apple II.

Eric Rosenfeld 15.437 presentation 2/19/09
http://techtv.mit.edu/collections/15437/videos/2450-eric-rosenfeld-15437-presentation-21909

THIS was indeed a remarkable presentation. Somebody who really made history, talking about his plan, his assumptions, and his entire way of thinking.
It's like having Gen. Erich Marcks explaining his planning of Operation Barbarossa.
Great stuff !
Thanks Paul !

michael webster 15p · 1 day ago

Very interesting, and nicely set out to counter some of the ordinary thoughts about LTCM.

Some of the social dilemmas which lead to the unfortunate and pointless increase in volatility would be fun to simulate.

For me, the important observation was the economic diversification means nothing if all your trading partners have similar risk preferences and decide to get out of your trades at the same time. I suspect this is similar to what happened in the sub prime contagion: previous uncorrelated housing down turns became one correlated market because of the risk preferences of who was holding that debt.

Mean Mister Mustard · 1 day ago

The comment at beginning about model doubters being "anti-intellectual" seemed a bit rich and that was the way much of the presentation felt. The Q&A seemed a lot more relaxed and thoughtful.

Josh · 1 day ago

Very thoughtful, but it seemed to me that he kept missing the obvious problem that his basic mathematical assumptions including what is sigma, were completely out of whack with the real world

GREAT stuff. Like we are there.

Rosenfeld does thankfully, however, have a whiff of the Mel Brooks humor/style about him! He gives good enthus. speech mon!
my takeaway: "it's good to be tha king" (history of world). A charmed life..burn a.billion here,...burn a.billion there....not our fault really.....who knew.....those damn russians.... (LOL). If these "riskless" trades blew up. what hope is there for hedge funds today.

all that MBA101 "portfolio theory" is out the door with separate counterparties eh !

Hubert · 17 hours ago

Rosenfeld comes across as a nice guy but I was expecting someone smarter.

Searching for 10 uncorrelated trades (as hidden principle, "trade secret") put them into some pretty stupid ones. (Merger arb, Volkswagen ).
Maybe it is just terminology but calling sigma "risk" makes one think Rosenfeld never left the academic orbit. He was with Salomon! Should have learnt the difference there.
Also that the umbrella of an investment bank has some advantages compared to a hedge fund - he took some time to admit to that.

Just one remark around the 70:00 minute makes me wonder: Starting 2007 Lehman had swaps wrapped with a AAA guarantee from AIG. This looks like a very interesting trace. Rosenfeld saw AIG as an artificial clearing house. But, more interestingly, it might indicate that other banks were no longer comfortable to do business with LEH back in 2007 - otherwise why would they go for wraps ? Has anybody any idea what the wraps were about and how they played into September 2008 ?

You'd think after 10 years he'd have gotten over the denial phase. This is truly disgusting.