Chairman and CEO of Olsen, Dr. Richard Olsen holds a Licentiate in Law from the University of Zurich (1979), a Masters in Economics from Oxford University (1980) and a Ph.D. from the University of Zurich (1981). He worked in banking as a researcher and foreign exchange dealer before founding Olsen in 1985 and becoming the firm's Chief Executive Officer. He is also CEO of OANDA.
Richard Olsen is involved in several companies, OANDA, OFT (Olsen Financial Technologies), Olsen Ltd and Olsen Investment Corporation. The companies have one theme in common - they contribute to his underlying objective of transforming finance into a 'high tech' industry and making it into a true engineering science with the same standards of quality and sophistication as the most advanced technology in the computer or pharmaceutical industry. For a number of reasons, modern financial markets have failed to evolve and continue to rely on primitive engineering tools. To illustrate this point, risk management in banking continues to use daily data for its risk computation, even though we know today that the daily computations have a large error term. Or another example, in the hedge fund industry a large fraction of the investment managers rely on traditional technical analysis for their trading decisions. Currently, most of my time is dedicated to Olsen Ltd and Olsen Investment Corporation. Olsen Ltd develops quantitative statistical trading models for trading currencies. Olsen Investment Corporation offers managed currency accounts, markets a hedge fund called High Frequency Data Fund and other currency related investment products.
 |
Every morning, Richard Olsen hops on his bike, buckles on his helmet and glides down the hill to his office in Zurich, Switzerland. In the evening, he pedals uphill to his home. "It takes about eight minutes to get to the office," he says, "but I need 16 or 20 minutes to get home - depending on the weather."
In his head, however, the uphill journey takes many times longer than on his watch. "That's the difference between intrinsic time and physical time," he laughs, referring to one of the epiphanies that got him hooked on "high-frequency finance" - his term for an approach to markets that, in part, measures time in terms of volatility instead of seconds. "It's the same concept that's behind point-and-figure charting, except we do it mathematically instead of visually, which is not an insignificant achievement."
Olsen heads Olsen Ltd. (www.olsen.ch), which is a combination hedge fund and open research project that has attracted some of the most inquisitive minds in finance - among them Benoit Mandelbrot, whose 1967 article "How Long is the Coast of Britain?" launched the field of fractal geometry.
Olsen's trading platform (www.oanda.com) uses an automated market-making engine developed by his Switzerland-based team of physicists and a Canada-based team of programmers, headed up by childhood friend Michael Stumm.
"We want to create tools of finance that are as slick and elegant as the most sophisticated tools of technology," he says. Toward that end, Olsen Ltd. operates a feature called a BoxOption, which allows traders to define a specific strike range in time and price with a mouse click. The system then hedges the position automatically, but trades are limited to just $200 at this point.
Olsen and Michael Stumm began trading stocks in their late teens. Then Olsen went to Oxford to study law and economics, while Stumm went to Stanford for post-doctoral work. In 1974, at the age of 21, Olsen read about a government subsidy to save jobs, and he realized the headline could as easily have said the government bought 10,000 jobs, with the salary of those people as the price. "That got me thinking about what would happen if the government capitalized unemployment so people could invest in new jobs," he says.
That launched his nimble mind on a series of inquiries such as why some things become assets while others don't, and how come only some of those assets become tradable. That no one would listen to his ideas about flexibility in the job market got him wondering how many other obvious concepts were being missed.
Two years later, he took a course in growth modeling taught by James Mirrlees, who 20 years later won the Nobel Prize in Economics for his work on the economic theory of incentives in a world where decision-makers have varying information. "I was very impressed by the fact that he told us up front that there was something wrong with the growth models we were learning," Olsen says. "Basically, the whole system broke down whenever one aspect broke down, while real systems are more self-adaptive."
After two years of trying to crack the problem, Olsen experienced another epiphany: "In most models, the building blocks were static, and dynamism was introduced as being something that happens to them," he says. "For models to be realistic, we had to make sure that the basic foundation was dynamic."
Part of that foundation, of course, is time itself. "Sometimes time goes by quickly, and sometimes it drags on and on," Olsen says. "When you see the market collapse 5% in a half-hour, for example, physical time has little meaning."
All these obsessions came to a head during a brief stint as a bank trader, when Olsen discovered technical analysis. "It was fascinating to see these ideas in practice," he recalls, "but it was also disconcerting to see that the raw data used to construct these precise models often included bad trades and unfilled bids and offers."
So, in 1985 he set up the company that became Olsen Ltd. and began recruiting physicists to help filter and structure the data. That brought him into contact with Mandelbrot, who now serves on the board of Olsen's Center of Emerging Finance, which conducts conferences on high-frequency finance.
By Benoit B. Mandelbrot and Richard L. Hudson
Wall Street, as ever, confounds. But the mysterious behavior of financial markets attracts academics eager to uncover investing secrets. For the past 40 years, Yale University mathematician Benoit B. Mandelbrot, founder of a branch of mathematics called fractal geometry, has applied his academic theories to financial markets. His findings are explored in a book he has written with Richard L. Hudson, former managing editor of The Wall Street Journal Europe, titled "The (Mis)Behavior of Markets: A Fractal View of Risk, Ruin and Reward." In the following excerpt, Richard Olsen, a Swiss fund manager and specialist in mathematical finance, illustrates Mr. Mandelbrot's view of the financial markets: They aren't mysterious, but rather physical systems that ought to be examined scientifically and engineered rationally.
Zurich -- THE NO. 4 STREETCAR heads from the city center, down the eastern lake shore, eventually arriving at the Mill Museum, a four-story, century-old factory now housing worthy exhibitions on cereals, the food industry and the age-old human cycle of famine and surplus, boom and bust. Next door, however, is a kind of laboratory for boom and bust -- a test reactor, its founder calls it. "What we're doing is quantum theory for finance," Richard Olsen says.
His company, Oanda.com, looks like just another small financial house. Barely 25 people man its market-making screens, trade e-mail with customers or work its computers. Its Web site, on foreign-exchange markets, seems humdrum. It has instant currency converters, live quotes, news, scholarly articles on market theory, trading games, downloadable software to analyze the market, and -- now something out of the ordinary -- a service that lets you bet real money on currency rates. If you open an account, you get what looks like a front-row seat at a forex dealer's trading screen. On your PC, you can chart the dollar/yen or euro/sterling market, project price movements, work out a trading strategy and then place a bet, with real money. It can be as little as $1. Launched in 2001, the service in early 2004 had about 10,000 customers with trading accounts. Most were amateurs, taking a flutter. But Oanda also attracts some big money. All told, its customers trade about $1 billion of yen, euro, dollar, bhat or pesos a day.
It is, in short, a small-scale model of the real currency market. One problem with almost all economic or financial research is limited information. If you want to study a market, you can get lots of generic numbers -- indexes, price quotations, volume. If you are inside a brokerage house, you can supplement that with precise information about what your own clients are doing, and, to some extent, why. But you can never see what other firms' customers are doing. You can never get the whole picture, the satellite view. That is what Oanda.com provides Mr. Olsen and his handful of math and finance Ph.D.s: the insight, both general and particular, of what people actually do in a market.
"I have this terrible sense of frustration," Mr. Olsen says. "We send space shuttles into orbit; we send probes to Mars; but we haven't studied the financial markets. We literally know nothing about how economics works. I want to break that deadlock. I want to change financial markets into something as efficient as engineering."
Mr. Olsen is a painfully earnest, lanky 51-year-old with a manner more suited to scholar than trader. In the world of forex, where he is well-known among the big bank research departments, he is viewed as something of a boffin: brainy, dedicated and perhaps a bit eccentric. He has a master's in politics and economics from Oxford and a Ph.D. in law from Zurich and worked among the financiers of Zurich. But he quickly became a prophet for an important new faith in financial research: high-frequency data. A century ago, even yearly data on broad trends were hard to come by. Then reporting of monthly, weekly and daily prices improved on exchanges and in newspapers. But the real data stream is tick by tick, quote by quote, transaction by transaction -- and that was available only in a few places, such as on the New York Stock Exchange. So in the 1980s, news services like Reuters began to see some value in transmitting instant-by-instant numbers to paying clients -- and that is where Mr. Olsen and his colleagues in Zurich saw opportunity. They amassed, debugged and began studying what has become one of the world's biggest databases of tick-by-tick foreign-exchange quotations. For academics, it has been a boon; scores of scholarly finance articles have been published based upon it. But the big banks, to which Mr. Olsen also hoped to sell use of the database, weren't much interested; his firm was liquidated.
Oanda.com was his next idea for studying the market, which he founded in 1996 with a school friend and computer-science professor, Michael Stumm. And it has been an entirely different story. In 2003, according to its reports to the U.S. Commodity Futures Trading Commission, Oanda's net capital more than doubled to $4.1 million -- a tidy profit. Olsen Investment Corp., a sister company, manages some relatively small sums -- euro 30 million ($36.3 million) at the end of 2003 -- for customers in the foreign-exchange market. The funds have performed fairly well. In 2003, the best fund returned 21.05%, the worst, 3.15%, according to audited reports. The performance difference from one fund to the next arises mainly from how much risk, or leverage, each fund tolerates; as is common in forex, the riskiest funds have done the best -- so far. But the trading strategy for all the funds is the same, and follows Mr. Olsen's computerized, quasi-fractal models of the market.
To him, a financial transaction is like a small explosion. Conventional financial theory, as taught in business schools around the world, holds that prices change continuously, and that each investor is as unimportant as the next. Their trades are like the collisions of molecules in a gas chamber -- millions of tiny energy exchanges. Nonsense, Mr. Olsen says. His tick-by-tick data show plainly that prices jump. Quotes stutter. And investors vary greatly in importance and impact on the market. A more accurate metaphor is the chamber in an internal combustion engine: Millions of small and large explosions drive the car forward, as the sparkplugs fire and the pistons churn.
As he sees it, in a well-functioning market small investors behave much like big investors, and make profits that scale proportionately. Only the industry's unfair commission structure and other idiosyncrasies tilt the game. Likewise, short-term traders act much like long-term investors -- again, with measurable scaling factors. He can see this, he says, in the computers tracking his Web service, FXTrade. There, fees are abolished and interest is compounded second by second; big and small investors are on an equal footing as they place their currency bets. To keep the system real, Mr. Olsen is registered as a market-maker, like the behemoths that rule the real currency markets. Mr. Olsen's computers keep his own quotes in line with those of the big banks, and also buy or sell real currency contracts to manage his own risk. Like other market-makers, he earns money on the spread, or the difference between the rates he sets to buy and sell a currency. But to those using his system, all that is invisible: What they see is just a currency market, and they can trade in it as often as they like, with whatever strategy or investment they like.
Mr. Olsen's fractal notions boil down to a theory he calls "heterogeneous markets." Orthodox economics is all wrong, he says. People aren't rational, and they don't all think alike. Some are quick-trigger speculators who pop in and out of the market hundreds of times a day. Some are corporate treasurers, deliberately buying or selling big contracts to fund a merger or hedge an export risk. Some are central bankers, who trade only occasionally, and at critical moments. Others are long-term investors who buy and hold for months or years. Each one, operating on his own time scale, comes together at one moment of trading, like all of time compressing into an instant, or the entirety of a rainbow spectrum focusing onto one white point. That is where the multifractal analysis comes in, he says: It is a mathematical tool for decomposing the market into its different elements and seeing how they interrelate and interact. And it suggests some real-world trading strategies. Using his models, his computers look for moments when the short-term traders are moving opposite to the long-term investors -- and then he bets that the imbalance will correct itself.
In the end, he says, his goal is to make the financial system work better and more safely. If the real market worked like FXTrade, costs would come down, liquidity would rise. "The world economy is like your body," he says. "Your heart pumps six liters of blood a minute, and so if you weigh eighty kilos it would take about fifteen minutes to pump your body's weight. By that analogy, the world foreign-exchange market should be transacting $40 trillion every 10 minutes. Today we do $1 trillion or so in 24 hours. My claim is the global economy is close to a heart attack."
The content of this site is copyright 2016 Financial Spread Betting Ltd. Please contact us if you wish to reproduce any of it.
