Paul Wilmott Wants To Retrain and Reform Wall Street's Quants 198
theodp writes "What if an aeronautics engineer couldn't reconcile his elegant design for a state-of-the-art jumbo jet with Newton's second law of motion and decided to tweak the equation to fit his design? In a way, Newsweek reports, this is what's happened in quantitative finance, which is in desperate need of reform. And 49-year-old Oxford-trained mathematician Paul Wilmott — arguably the most influential quant today — thinks he knows where to start. With his CQF program, Wilmott is out to save the quants from themselves and the rest of us from their future destruction. 'We need to get back to testing models rather than revering them,' says Wilmott. 'That's hard work, but this idea that there are these great principles governing finance and that correlations can just be plucked out of the air is totally false.'"
You can't blame it all on the qunats. (Score:5, Informative)
Quants only produce models that act in the way that traders expect, and traders do not want bad news. I've done a small bit of modelling before and you always reach a point where there's this one number that is completely made up, and you kinda set things up so the trader makes the call. In this sense, all these models that everyone talks about are not so much as analysis tools as they are communications tools - you sorta code the insight of the trader as to how he or she thinks the market will move. It's a very human business, not one of a bunch of computers run amok. Quants that say otherwise are just full of themselves...
Re:how about reforming pay? (Score:2, Informative)
Do you mean Long Term Capital Management?
Founding members included Nobel prize winners Myron Scholes and Robert C. Merton. Imploded spectacularly in 1998.
http://en.wikipedia.org/wiki/LTCM [wikipedia.org]
Re:How about... (Score:3, Informative)
Nice job [newsweek.com] Newsweek (someone take a screenshot before the article gets pulled/corrected). His hedge fund didn't even exist in 2001 [caissacapitalltd.com]. And it's not a $170 million fund, it's roughly a 170 million British Pounds [guardian.co.uk] fund. I would add more, but his hedge fund was delisted from the New York Stock Exchange and the London Stock Exchange in January 2005 right after this fiasco [timesonline.co.uk] (and it's funny, except for one cryptic note [caissacapitalltd.com] about taking counsel in their official news section, one wouldn't know that the fund got liquidated and delisted four years ago for its shenanigans).
Re:Wow! (Score:5, Informative)
You are right, this is one important source of problems. I started out in quantitative finance back in the 1970s. (I worked as a research assistant for Fischer Black when I was in grad school.) The initial application of many of the quants' techniques were in markets like US equities, or listed options, where the assumptions that one participant couldn't affect the overall market much and that there were reliable sources of information were probably reasonable.
But if you look at one of the key "villains" in this last mess, the credit-default swap [CDS] market, it's an entirely different story. I have read Li's paper on the Gaussian copula function, and had a look at an implementation. What it is essentially doing is using a statistical sampling function to estimate the expected lifetime to failure (= default) for a population of debt instruments. Now, there is nothing wrong with the math per se; similar approaches are used in manufacturing for quality assurance. However, there is big difference: estimating the failure rate of, say, light bulbs does not in itself have any effect on that rate. But in the case of the CDS, the failure rate is being used as an input to the model that is used to price the swap. If the default rate estimate is too low (too optimistic), the prices will be too high -- and that, in turn, will lead to lower estimates of the default rate. In essence, there is a built-in feedback mechanism that can act as an error amplifier, a problem that is exacerbated by the lack of transparency and liquidity in the CDS market.
There's plenty of blame to go around. The managements, who should have known better, were bedazzled by the dollar signs floating out of their economic perpetual-motion machine. The quants knew the math, and their hubris led them to think that nothing else was needed. And the investors, while proving the truth of P.T. Barnum's Law of Applied Economics, forgot that there ain't no free lunch.
Re:How about... (Score:4, Informative)
Ok, the link didn't work. I'll try again:
December 9, 2004
Caissa founders in court battle
By Joe Morgan and Richard Irving
A FORMER Oxford University professor, a chess grandmaster and a software developer are locked in a court fight in New York over ownership of a £172 million hedge fund in which top City names such as Gavyn Davies have invested tens of millions of pounds.
Paul Wilmott, Ron Henley and Jonathan Kinlay, who set up a secretive hedge fund, Caissa Capital, accuse each other of trying to lure top investors to rival funds after the partnership collapsed in acrimony earlier this year.
The dispute erupted in September when Mr Kinlay, who has developed a computer program that highlights profitable trading strategies in volatile markets, filed a lawsuit in the Supreme Court of the State of New York against his two former partners. Mr Kinlay is seeking $800,000 (£414,000) in damages for alleged breaches of intellectual property rights.
Mr Wilmott, a former Oxford University professor, and Mr Henley, who trained the former chess world champion Anatoly Karpov, hit back with a $10 million counter-suit, claiming that Mr Kinlay had tried to wind up Caissa by luring investors to a new fund that he was secretly trying to set up without their knowledge.
The dispute flared up after Mr Kinlay returned from a holiday in Italy in August to find that Mr Wilmott and Mr Henley had given back $40 million to Caissa's biggest investor, Prisma Capital Partners. Prisma is run by Gavyn Davies, former Chairman of the BBC, Girish Reddy, former head of derivatives for Goldman Sachs, and Tom Healey, a star banker.
Mr Wilmott and Mr Henley claim that Prisma demanded the return of its money after Mr Kinlay approached its backers with a proposition to invest in his new venture, the Proteom hedge fund. It is understood that the partners were sceptical of the strategy Mr Kinlay planned to use, which essentially involved using the same technology that looks for patterns in genes to predict stock and bond market moves.
Mr Kinlay, meanwhile, alleges it was his two former partners who were trying to lure Prisma into a new fund that they were also trying to set up. Both sides deny the allegations.
After the initial dispute, the trio decided to go their separate ways and arrangements were made to write a separation agreement in August. On the basis of this agreement, Mr Kinlay agreed to sell his interests in Caissa Capital and Caissa Capital International, an overseas offshoot, to Mr Wilmott and Mr Henley. He also agreed to terminate software licences granting access rights to his trading program. He collected $377,759 from his former partners under the deal.
However, court papers allege that the separation agreement negotiated by Mr Kinlay was just a smokescreen while he asked a bank, which was helping him to set up the Proteom fund, to find a legal "rottweiler . . . who would sue his two partners and Caissa's investors".
It is alleged Mr Kinlay began to plot against his former partners almost a year earlier after deciding that neither "had the time nor the skill set to make a meaningful contribution".
Court documents allege that Mr Kinlay set up the Proteom fund because, as a minority partner in Caissa, he could not force Mr Henley and Mr Wilmott out. Mr Kinlay denies the allegations.
PARTNERSHIP BREAKDOWN SEEN AS CHESS CONFLICT
Page 1 of 2
Re:How about... (Score:4, Informative)
Here is the second page:
"There is nothing that would persuade me to remain in a partnership with someone as stupid, duplicitous and untrustworthy as you have proved yourself to be. As we discussed, I shall make arrangements for your exit from this firm at the earliest convenient opportunity."
Jonathan Kinlay to Ron Henley, July 20, 2004
"How like a game of chess this is. Except you are toying, not with wooden pieces, but with people's lives. And you are about to discover in this game it is I who am the grandmaster."
Jonathan Kinlay to Ron Henley, July 25, 2004
"White's attack has been successfully put down. Black sacrifices the terminally weakened pawn in order to open lines for the coming counter-attack. Meanwhile, the white knight has been neutralised and lies isolated and vulnerable at the edge of the board, waiting to be picked off by the black forces at their leisure."
Jonathan Kinlay to Ron Henley and Paul Wilmott, August 15, 2004
THE BRAIN'S WHO FELL OUT
JONATHAN KINLAY
The chief executive of Investment Analytics, a mathematical research firm that develops software programmes to exploit volatile stock markets.
He started his career at NatWest in the early 1980s but had left long before the investment bank found an £80 million black hole in its options trading book.
After a spell at Chase Manhattan, he joined the proprietary trading desk of EMC International, a European hedge fund, specialising in privately negotiated derivatives contracts. He is well known on the lecture circuit and has taught financial engineering at Carnegie Mellon in New York and at Oxford and Cambridge.
RON HENLEY
Few people span the diverse worlds of chess and high finance, and fewer rise to the top of both. Ron Henley is a chess grandmaster, but he is best known for training Anatoly Karpov, the former chess world champion. His interest in derivatives dates back to 1985, when he became a member of the American Stock Exchange.
He soon earned himself a reputation as a derivatives whizz kid, rising to become a specialist options trader for Cohen, Duffy & McGowan, one of the top options markets makers on the exchange floor. He is a regular chess commentator and runs an internet site that raises funds to sponsor young US players, including Irena Krush.
JONATHAN KINLAY
The chief executive of Investment Analytics, a mathematical research firm that develops software programmes to exploit volatile stock markets.
He started his career at NatWest in the early 1980s but had left long before the investment bank found an £80 million black hole in its options trading book.
After a spell at Chase Manhattan, he joined the proprietary trading desk of EMC International, a European hedge fund, specialising in privately negotiated derivatives contracts. He is well known on the lecture circuit and has taught financial engineering at Carnegie Mellon in New York and at Oxford and Cambridge.
Page 2 of 2
Re:You can't blame it all on the qunats. (Score:3, Informative)
...you always reach a point where there's this one number that is completely made up...
This is true, but your proposed methods do not eliminate this. Yes, a sensitivity analysis can help. But the only advantage of prior distributions over parameters is that they encourage one to put all their assumptions on the table, whereas frequentist statistical methods (fixing parameters) tend to hide things. Other than that, you will always be subject to your modeling assumptions.
***Try a sensitivity analysis using Monte Carlo techniques. That sounds hard but it isn't.
Yes, it's easy to just "do"; doing it correctly in a way that never gives you false information or gives you accurate confidence bounds can be extremely difficult. Not that it doesn't work a lot of the time, but there are dozens of gotchas that can cause the answer to be complete rubbish and no one would know without a lot of very careful math and analysis. Yes, it can be better than other methods, but only if used properly.
Take the parameter that you have doubts about and give it a distribution (Gaussian or rectangular or something) with the mean at your best guess and the std deviation chosen to be big enough to cover the range it might reasonably vary over.
Um, yes, and those parameters have a big influence on your results... For example, if you center your prior on the parameter you think it is, it is generally NOT true that the mean you get out will be even close to the value when just plugging in that parameter. Most real data in industry and finance is not subject to the natural processes that seem to turn most things Gaussian.
***Use Gaussian if you have an idea of what the parameter probably is but aren't exactly sure, rectangular if you really have no idea. A rectangular distribution says "I have no idea, the parameter could be anywhere within this particular range."
HaHAHAHA. Can I quote that next time I teach? A bounded uniform prior, when done with Monte Carlo, often denotes MUCH stronger assumptions than does a Gaussian or t-distribution. It is basically say, "there is no chance whatsoever the parameter can be out of this range." So, you say, make your rectangle large enough. Well, that only works for undergrad stats courses, not in most of the models I've worked with or dealt with. It also breaks down phenomenally fast in higher dimensions. It may work as a hack, but I would NEVER trust such results unless there is good reason to use a bounded uniform.
****Monte Carlo is great for those of us who don't care to learn the arcane minutiae of stat math. If you have a working model it takes an hour or two to extend it so you get stochastic results. Note that it's no harder to give a distribution to all your input parameters not just one. In which case you will be doing the kind of work that people who make 500 grand a year do.
If you don't go through the math and simply treat it as a black box, you WILL MESS monte carlo up and give false results. We need more people in that sector who really know statistics, which means mastering statistical math (and I'm curious why you think it's arcane), not just think they do and plow blindly through minefields of gotchas you never learn in undergrad stats courses. Yes, MC is a great tool; it may be a step up, but I would never trust it without having good theoretical justification that it works. On the models in the financial industry, this is much more difficult than you might think.
See sig!!!