|
|
|
|
 |
Daily Speculations The Web Site of Victor Niederhoffer & Laurel Kenner Dedicated to the scientific method, free markets, deflating ballyhoo, creating value, and laughter; a forum for us to use our meager abilities to make the world of specinvestments a better place. |
Write to us at:
(address is not clickable)
Quantitative Trading Strategies: Harnessing the Power of Quantitative Techniques to Create a Winning Trading Program by Lars Kestner
Reviewed by Duncan Coker
Quantitative Trading Strategies is a thoroughly researched book in which the author looks at thirty different markets over a 10 year period from 1990-2001 and applies numerous trading strategies to each and reports the results. To generalize, the results show that trend following strategies performed very well in most markets, include energy, fixed income, commodities, certain basket of stocks and foreign exchange. The notable exception was SP futures which did not perform well using most momentum type trades.
The book is broken out in to four sections; the first explains the quantitative measures he uses, the second is analysis of a series of trades on various markets, the third does the same analysis but on relative/pairs trades, and the last looks at certain specific pricing anomalies and explains money management and trade optimizing techniques.
The book has a lot of information and some good ideas, but spends too much time on technical trades that rely heavily on moving averages and lagging data. Most of the central trade ideas are based on technical indicators, like channel break outs, moving average convergence divergence, stochastic and oscillators. On the positive, he does a thorough job of defining and quantifying exactly what these are, and reports the results statistically for easy comparison. He uses risk measures like standard deviations and regression variance scores.<
Most of the system trades are momentum driven, buying high and selling the low. In the results sections he shows the aggregate for all markets for each system along with Sharpe ratio and K-ratio both for individual markets and the portfolio as a whole. K-ratio is defined as the slope of the regression line, divided by the standard error times a scaling factor. The channel breakout (buy x day high, selling x day low) is the core trade idea along with moving average cross-overs and buying when today is higher than x days ago. In addition he does include some mean reversion strategies which reversing a recent move up or down. Other trades involve and constructing bands around moving average indexes one or two standard deviations and buying or selling the breakouts.
The interesting part of the results is how well many of these markets did during the period. Bonds and foreign exchange and energy made of the majority of profits for these types of trade and more than made up for the loses experienced in SP futures. Individual stock categories like tech also did well. However, most of the trades testing showed decreasing performance, making most of the profits in the early 1990s and less so as the strategy ages. In the relative value-pairs trading section he does a nice job of explaining and defining the various markets, including statistical arbitrage, credit spreads, equity volatility and yield curve spreads. When he applies that same trade ideas to the relative spreads trades the results are almost reversed. Markets like volatility, energy spreads, and individual stock momentum strategies lost money because relative values showed mean reversion prices action. Yield spread and stock index spreads benefited from momentum trades on a relative value basis.
That last part of the books outlines some specific trades the author found beneficial like trading the end and beginning of calendar months. By just trading these periods, $100 increases to $454 versus $419 for just buy and hold the whole period from 1990-2001. Another trade is designed around multiple standard deviation moves in the VIX away from its moving average. This outperformed buy and hold over the period with half the market exposure.
He also tests the use of stop on trades and finds that it improves results on channel breakout momentum trades by a factor of almost two. In one of the most valuable sections of the book he calculates a regression equation between bond yield and stock changes and highlights the strong relationship between the two over the period from 1970-2001. A chart illustrates 37 different trades, most using bonds as a trigger for entering stock trades. The most profitable trade was buying the SP when the 10 year yield is below it 3 month moving average and exiting on the opposite. This strategy turns $100 in 1970 to $6819, versus $3661 for buy and hold.
I give the author much credit for his work. However I would not recommend the specific trades suggested in this book for equity or equity index markets. The profitable trades were mostly in the bonds, energies, commodities and foreign exchange. Also many of the momentum profits were breaking down over the period and may not work in the future. The book does provide a good outline for doing one s own research and coming up with new ideas. Also, the authors constructs as excellent view of all markets during the last decade for easy comparison to make general conclusions. And for those followers of technical indicators, this is the most quantified study I have seen on how they perform and when and where they are useful.