Author: Eric Uhlfelder

18 May 2009, Barron’s

Roy Niederhoffer has been able to generate consistently strong returns that are negatively correlated to both stocks and hedge funds by focusing on the ultra short-term behavior of investors.

Few may see a connection between Wagner’s Ring cycle and investing. But Roy Niederhoffer does.

An accomplished performing classical musician, Niederhoffer recognizes the compositional structure in Wagner’s operas in which repeating themes foretell musical passages that are to follow. And he has found a similar pattern in the way investors respond to changes in the price of stocks, currencies, and commodities.

Applying this thinking through a high-frequency trading strategy that invests across 55 financial and commodity markets throughout the world’s developed economies has produced remarkable results. Over the past year through March 2009, Niederhoffer’s Diversified Trading Program, which ranked 9th in Barron’s recent top 100 hedge fund survey [Acing a Stress Test, 11 May 2009] was up 34.58%. It topped the Barclay Hedge Fund Index by more than 52% and S&P 500 by more than 70%.

The program, with $480 million in assets as of the end of April, easily blew past most of the competition over the past 5 years with annualized returns of 13.7% versus the hedge fund industry average gain of 2.51% and an annualized broad market decline of -4.76%. And since its inception in 1993, Niederhoffer has realized annualized returns of nearly 12 percent, more than twice the S&P’s average gain. [BarclayHedge’s data for hedge funds doesn’t go back that far.]

How has he achieved such a trading record? Through extensive study of historical trading patterns, tracking 10 to 1,000 trades every hour of the day, Niederhoffer found that over the short term when markets behave in a certain way, investors tend to respond predictability.

That behavior may be ephemeral. But that’s just fine to Niederhoffer because unlike longer-term investors, his trading programs target quick gains. Last year he made about 12,000 trades, lasting between several hours and a week. And being right just a little more than half the time is all he needs to generate large profits.

When stocks sold off in the aftermath of the Lehman Brothers bankruptcy or the AIG collapse, recalls Niederhoffer, his models strongly suggested a short-term rebound was very likely in both instances. They were right.

If this approach sounds like it ignores fundamentals in favor of more esoteric analysis, it shouldn’t surprise. With a background in computational neuroscience, Niederhoffer is used to focusing on the psychology of investors rather than on corporate earnings and macroeconomic growth rates.

He contends that short-term investor behavior derives from several basic tendencies all hard-wired into the human brain: one, investors overemphasize recent experience when making decisions; two, people hate to lose more than they love to win; three, investors become more predictable as markets become more volatile; and four, when people are emotional, they tend to behave herd-like [i.e., selling at the worse times or buying into overextended rallies].

Not only has this approach succeeded in producing consistently impressive returns, it’s done so in a way that’s most important to sophisticated investors: negative correlated returns. Last year, the Diversified Program had a correlation of -0.68 to the S&P 500 and -0.74 to the Hedge Fund Research Global Hedge Fund Index.

Based in New York with a staff of 41 and currently managing more than $1 billion, Niederhoffer set up his firm in 1993 as a commodity trading advisor. Through his flagship Diversified Trading Program, he began investing in major financial and commodity markets following just a handful of trading rules.

But after two particular years of poor performance in 1995 and again in 1999 as stocks soared, he realized he needed to substantially broaden his then purely contrarian focus. “Especially at the peak of the Internet-based bubble in 1999 when the market was up a third and we were off by nearly 20%,” Niederhoffer recalls, “we found all our hulls were breached and that to prevent us sinking, we needed greater diversification.”

So he introduced a series of investment strategies [families he calls them which now number 8] that balance contrarian positions with momentum plays of various durations. He also drastically increased the number of trading rules [from 10 up to 60] supporting these strategies, each geared to a distinctive bias that forecasts reaction to recent market behavior.

Unusual for a CTA, he added individual stocks to the portfolio, picking from a universe of 800 US and European equities to further enhance diversification over straight equity index exposure.

The result drastically improved the program’s performance. From inception through 1999, annualized returns were basically flat. Since then through March 2009, annualized performance exceeded 17 percent. Especially important is the fund’s ability to consistently generate gains with less risk. Here’s how he does it.

Niederhoffer, who wrote the original computer code to test and run his trading rules, designed the Diversified Program to follow a variety of signals that search out uncorrelated market positions. He can increase emphasis on short-term momentum that can counter positions that are getting hit. Restricting exposure to both individual asset classes and to particular trading rules prevents the program from getting hurt from a particular bias that could go wrong.

The program is calibrated to reduce leverage and risk when Value at Risk [the amount the program could loose in any one period] exceeds certain thresholds. The firm’s risk management committee can step in to reduce risk at any time, sometimes through buying options to protect larger positions. And the program will apply additional defensive measures when a monthly drawdown exceeds 8%. Collectively, these features help limit average daily volatility to 1% and annual volatility to 16%.

This past March 2009, these measures limited a loss Niederhoffer incurred when he boosted his net long position on US Treasuries from 5% to 30% after the Fed announced it was purchasing Treasuries to pump capital into the economy. This apparent commitment to quantitative easing sent US 10-year Treasury prices soaring 4% in one day. But instead of continuing to rally, Treasuries suddenly sold off nearly 75 basis points over the next two days. The firm’s model triggered a sell, limiting the loss to 1%. It could’ve been worse: Treasuries subsequently slid by roughly another 125 bps.

But a majority of the time, Niederhoffer’s bets are in the black. In just several days last July, as commodity prices were in free fall, oil futures price collapsed from $145 to $136. At the time, the program was 15% net short oil and gas futures, which were moving in tandem. His screens then suggested that over the near term the selloff was excessive and to switch to a 15% net long exposure. Within 48 hours, oil had rallied back up to $140, at which point the system moved to a sell, netting the program a 2% gain on his investment.

Later in the fall, the government’s announcement of a temporary ban on short selling of financials on September 18 sent the market up nearly 12% as short sellers were forced to cover their positions. But soon after, the program’s contrarian models were indicating the market was extremely overbought, setting up a heavy 60% net short play in S&P 500 futures. By September 22, stocks sold off and Niederhoffer exited his short position with a 4% gain.

After falling against the dollar for most of January, the euro started to rally in the last week of the month, rising from $1.29 to $1.32. The firm’s screens suggested this was only a temporary reversal in the dollar’s rally. So the program went 25% net long the dollar. In two days, the euro fell back to $1.30, the position was sold, netting the program nearly 1%.

This year so far has proven rough for the Diversified Program, which is down 1.9% through April. According to Niederhoffer, this reveals the difference between an equity trend follower, who could’ve ridden the rally, and his ultra-high frequency trading strategy, which profits when price movements are constantly reversing course.

But having missed out on one of the great rallies doesn’t concern Niederhoffer because he likens it to a 100-year flood. “You can’t design a program around this very rare event, and you certainly don’t readjust your strategy to it.” Given the underlying challenges plaguing the global economy, he suspects that the current rally will run out steam, with markets returning to their more normal volatile state, which will be just fine to Niederhoffer.


Leave a Reply