Numerous public statements from Simons that are either demonstrably false, misleading, incomplete, contradictory, or just don't make any sense. Numbers surrounding the Renaissance fund that don't add up. A monthly return series that shows stong evidence of either return smoothing or unexploited structure. I vote for the smoothing, especially since a near-optimal universal portfolio of "small and ephemeral edges" (Simon's words) should show more month-to-month variability. There is something wrong with the Renaissance story. It rings only slightly more true than Eddie Lampert's kidnap story. Axcom was renamed Medallion. They are the same fund. The graph shown on this page: http://math.berkeley.edu/~berlek/fin.html used to be titled "Renaissance Medallion Fund." Berlekamp took off the title earlier this year after I posted about the connection on another forum. He did mention the connection in a widely read review of "Fotune's Formula" for American Scientist. Here is the link: http://www.americanscientist.org/te...etail/assetid/47321;jsessionid=baa9OLgyl1RbGr read the last sentence of the review. Berlekamp has a background in Information Theory. His graduate advisor was Claude Shannon. He also worked closely with John Kelly, of Kelly Criteria fame. What he brought to Renaisance was the risk/position/allocation management, the volatility pumping third of the equation. Simons already had the statistical mining for small edges part, and he developed the perception management part over the years -- now he has that part down to a science! .
Yes, $1.5 billion is a nice payday. Too bad it doesn't add up. It is either too high or too low. Medallion fund has no outside investors. Simons gave that money back. ALL of the money in Medallion belongs to Simons, his current and ex-employees and family. If the $1.5 billion includes the earnings on his share of the fund, then the number is too low and should be over $2 billion. On the other hand, if it is only what he was paid by Renaissance, then it is too high. $1.5 billion is the entire management plus incentive fees combined for 2005 with no overhead (160 employees on LI and in Manhattan, 50 acre campus...etc.) taken out, no profit sharing, no ownership interest for even long term executives who have been with the firm 15 years... A more realistic figure for what he was "paid" is probably under $1 billion. Also, interestingly, Forbes has his wealth dropping by $100 million between this year and last. I guess they missed the Alpha survey. By my calculations his net worth should have increased by over $2 billion, even taking into account his considerable charitable donations. Either I am missing something, or Forbes is. .
Thx, Kevin! Very insightful responses. Yes I agree with you, I've been following Simons since 1994 and I too have noticed the funds that went bonk and the various contradictory statements. I wonder how the Berkeley prof feels about selling out his stake in Medallion/Rentec....
Great insight on Renaissance. Thanks, Kevin! Has the relevant article been removed from EB's site? Would you still have a copy, Kevin? http://math.berkeley.edu/~berlek/
thanks Kevin, i have looked up the Splus paper; not sure if i have found the right one (it was on the FX crosses), but it's nice to get at least a remote idea of what they are doing. would someone knowledgeable be kind to explain / summarize in layman's terms (or otherwise) what "volatility pump" might be? i have looked up Thomas Cover's book on amazon, but i am not sure that i have sufficient math background to fully comprehend it.
I also fail to see the connection between "volatility pump" and Thomas Cover's book on information theory. Can anyone explain?