Low-Key Quant Strategy Thrives in Choppy Market |
Date: Wednesday, October 3, 2007
Author: Chidem Kurdas, Hedgeworld.com
NEW YORK (HedgeWorld.com)—Watching the performance of some quantitative hedge funds this August was enough to increase an investor's blood pressure, so wild were the gyrations. There were funds that lost 30% in a few days, gained 20% in the next few days, then lost some more and ended the month in the red by double digits. Quant managers argue that such short-term volatility should not matter, that over time the computer models do well, and they point to big gains in September to bolster their argument. Quants do tend to fall in lock-step, but they also tend to recover quickly Previous HedgeWorld Story. Many investors prefer not to ride a roller coaster, however. And not all model-driven trading resulted in a rough ride, despite higher correlations among traders in July and August. There were quantitative approaches that did well amid market turmoil. What made the difference? That question intrigued Mark Abeshouse and Matthew Knapp, co-managers of Augustus MJK Fund, which pursues a quantitative global equity strategy. Mr. Abeshouse has identified several factors. Augustus Capital LLC, a broker-dealer in Harrison, N.Y., seeds hedge funds. The Augustus MJK Fund's August performance was surprisingly strong. The fund was not totally immune to the summer turbulence—it took some hits, but only a few and they were not big ones. "We didn't lose much because we have much lower leverage than many quantitative market neutral funds and a limit on our net exposure," explained Mr. Abeshouse. "We stepped out of just a couple of positions." Compared to aggressive quant strategies, the Augustus MJK Fund has a relatively modest target rate of return. The aim is to make 1.5 percentage points a month consistently, avoiding any big loss. Since the impact of a big loss compounds over time, this enhances long-term returns. "The best way to make money in the long run is to not lose it in the short term," Mr. Abeshouse said. Thanks to its less aggressive target, the fund does not need to be highly levered. Many statistical arbitrage funds lever their capital five to 10 times. By comparison, Augustus MJK does not go beyond 2-to-1 leverage. In September, it was levered only 1.7 times. "If you try to make 25% or 40% a year, you're necessarily more leveraged," Mr. Abeshouse said. "We don't look for outsized returns." The fund is usually not fully invested and has cash to deploy when opportunities arise in the market. The Augustus quantitative model is a "screen of screens" that analyzes other models. It screens 30 models, long and short, that identify outlier stocks. The screen comes up with a ranking of stocks worldwide. While based on this ranking, the investment choices are also subject to risk control and diversification rules. Trades tend to focus on global midcap stocks. Around 40% to 50% of the portfolio consisted of U.S. equities in recent months, but the proportion changes with the markets. Non-U.S. companies, particularly in sectors like energy and metals, may contain more pricing anomalies and hence more trade opportunities than companies in the more efficient U.S. market. But the international allocation does not appear to explain much of the fund's success in August. One intriguing pattern is that most of the above-market return, or alpha, comes from short selling. It may be that because so many people trade long-only, the market is more efficiently priced on the long side, Mr. Abeshouse suggested. As a result, the screen picks up more gains on the short side. Another source of profit is quick trading around the core positions, exploiting market runups and downward corrections during the day. Technical indicators are used for this day trading. These helped a lot in July and August, Mr. Abeshouse said. Now he's investigating another question. The fund made 3.7% in August. The 9.7% year-to-date return is on target, but the August return is comparatively too high. Is there a problem? Mr. Abeshouse said he and Mr. Knapp are reviewing the matter.
Reproduction in whole or in part without permission is prohibited.
