Portfolio Performance Measurement and Benchmarking by Jon A. Christopherson, David R. Cariño, and Wayne E. Ferson (McGraw-Hill, 2009) is a welcome addition to the literature on investment management. Although its intended audience is portfolio managers—institutional investors and hedge fund managers, the individual investor with basic math and statistics skills can profit from it as well. In fact, I intend to mine the book for useful nuggets now and again on this blog because the authors have done a masterful job of synthesizing research on performance measurement and their original work on benchmark construction is both thorough and enlightening. (Two of the authors are research fellows at Russell Investments; the third is an academic.)
Today I want to focus on the paradox of the efficient market because the authors offer an interesting twist on it. Basically the paradox claims that the efficient market hypothesis is true only if enough investors believe it to be false. If everyone believed in market efficiency no one would try to discover inefficiencies. “In such a world, price changes would slow to a crawl or cease because there would be no one who would believe that it is wise or prudent to pay anything other than the market price.” New information would not be incorporated into a stock’s price, so the market would become less responsive and hence less efficient.
And the twist on this paradox: “. . . in order for passive management to be a viable investment strategy, it critically depends upon a large number of individuals who do not believe that current market prices are efficient. As more and more money is invested in index vehicles, the markets tend to become less efficient. Yet inefficiency should make active investment more attractive and hence lead to a fall in the percentage of market capitalization invested in passive vehicles.” (p. 246)
Extracting two paragraphs from a 466-page book is, of course, a sorry excuse for a review, but take them as illustrative. This book is a treasure trove of insights for anyone interested in performance metrics, especially benchmarked performance metrics.
Thursday, August 6, 2009
Wednesday, August 5, 2009
Futia, The Art of Contrarian Trading
The first part of Carl Futia’s The Art of Contrarian Trading: How to Profit from Crowd Behavior in the Financial Markets (Wiley, 2009) is quasi-theoretical; it is a well researched, tightly written analysis of investment crowds. The second part is practical—how to make money being a contrarian. Since I’ve always been more attracted to the “why” than to the “how,” it’s no surprise that I found the first part of the book more compelling than the second.
Futia makes his case for contrarian trading by appealing to the “No Free Lunch principle.” As applied to investing, it means that nothing that is widely known—fundamental, technical, or statistical—will give the speculator an edge and help him beat the market. It also means that there is no financial edge in joining an investment crowd, however comfortable and socially rewarding such membership may be. In fact, “it is his ability to suffer the internal conflicts and the social isolation associated with a contrarian investment stance that is the source of a contrarian trader’s edge.” (p. 29)
Futia criticizes market timing, pointing out that even if the market timer is highly skilled and is correct 70% of the time, he has to be right twice in a row; the probability of this happening is 70% x 70%, or 49%. Yet the contrarian is by definition a market timer even if he uses different tools from the technician. “His business is exploiting market mistakes arising from the growth of an investment crowd but then stepping aside from his investment as the life cycle of the crowd inevitably returns price to fair value.” (p. 28)
The one deadly mistake a contrarian can make that will lead to underperformance of the buy-and-hold strategy is to be out of or short the market when it is rising. And yet novice contrarians tend to commit this very mistake. Futia suggests that since it is much more difficult to ascertain when the market is overvalued than when it is undervalued, the novice should focus his attention on periods of undervaluation.
How does the contrarian go about figuring out market mistakes? Futia relies mainly on his media diary (including magazine covers and headlines), secondarily on the distance of the S&P 500 from its 50 and 200 day moving averages. He offers strategies for both conservative and aggressive contrarians using mostly broad market ETFs.
Futia makes his case for contrarian trading by appealing to the “No Free Lunch principle.” As applied to investing, it means that nothing that is widely known—fundamental, technical, or statistical—will give the speculator an edge and help him beat the market. It also means that there is no financial edge in joining an investment crowd, however comfortable and socially rewarding such membership may be. In fact, “it is his ability to suffer the internal conflicts and the social isolation associated with a contrarian investment stance that is the source of a contrarian trader’s edge.” (p. 29)
Futia criticizes market timing, pointing out that even if the market timer is highly skilled and is correct 70% of the time, he has to be right twice in a row; the probability of this happening is 70% x 70%, or 49%. Yet the contrarian is by definition a market timer even if he uses different tools from the technician. “His business is exploiting market mistakes arising from the growth of an investment crowd but then stepping aside from his investment as the life cycle of the crowd inevitably returns price to fair value.” (p. 28)
The one deadly mistake a contrarian can make that will lead to underperformance of the buy-and-hold strategy is to be out of or short the market when it is rising. And yet novice contrarians tend to commit this very mistake. Futia suggests that since it is much more difficult to ascertain when the market is overvalued than when it is undervalued, the novice should focus his attention on periods of undervaluation.
How does the contrarian go about figuring out market mistakes? Futia relies mainly on his media diary (including magazine covers and headlines), secondarily on the distance of the S&P 500 from its 50 and 200 day moving averages. He offers strategies for both conservative and aggressive contrarians using mostly broad market ETFs.
Tuesday, August 4, 2009
Risk aversion, gambling with the house's money, and playing long shots
In 1990 Richard H. Thaler and Eric J. Johnson wrote a paper entitled “Gambling with the House Money and Trying to Break Even: The Effects of Prior Outcomes on Risky Choice” (available for download at the Social Science Research Network). The paper focuses on how risk-taking behavior is affected by prior gains and losses. It definitely has repercussions for the trader and investor.
There is no need to summarize the authors’ theoretical work here. But their experiments led to three important empirical results. First, how do people respond to the chance to gain back some of their money after a loss? For instance, the subject had just lost $30 and was asked to choose between (a) doing nothing and (b) a gamble with a 50% chance to win $9 and a 50% chance to lose $9. Sixty percent of the subjects opted to do nothing.
Second, if the subject is ahead by $30, how does he respond to the chance to add to his winnings with the same $9 50-50 bet? Seventy-seven percent wanted to take the bet. They were playing with the house money. “The essence of the idea is that until the winnings are completely depleted, losses are coded as reductions in a gain, as if losing some of ‘their money’ doesn’t hurt as much as losing one’s own cash.” (p. 657)
Finally, risk aversion in the face of a previous loss can be overcome if the subject is offered the opportunity to break even (about 70% were risk-seeking in this scenario), especially with a modest outlay of money. The authors point to the proclivity for losing gamblers at the race track to bet on long shots at the end of the day. Although a risk-seeking bettor who is behind by $30 “could bet $30 on an even money favorite as a method of getting even, . . . [a] $2 bet on a 15-1 long shot offers a more attractive chance at breaking even because it does not risk losing significantly more money.” (p. 658)
The implications for the trader are obvious. First, it goes against the grain to pull the trigger after a losing trade and, by extension, after a string of losing trades (even when following a well-researched system) it requires an almost superhuman effort to take that next buy or sell signal. Second, winning traders get cocky because they no longer see themselves as playing with their own money. How many times have we read “move your stop to break even, then you’re playing with the house’s money.” The third result has two incarnations: the revenge trader who throws caution to the wind in his attempt to break even and the penny-wise and pound-foolish trader who is seduced by the siren call of out-of-the money, long shot options.
There is no need to summarize the authors’ theoretical work here. But their experiments led to three important empirical results. First, how do people respond to the chance to gain back some of their money after a loss? For instance, the subject had just lost $30 and was asked to choose between (a) doing nothing and (b) a gamble with a 50% chance to win $9 and a 50% chance to lose $9. Sixty percent of the subjects opted to do nothing.
Second, if the subject is ahead by $30, how does he respond to the chance to add to his winnings with the same $9 50-50 bet? Seventy-seven percent wanted to take the bet. They were playing with the house money. “The essence of the idea is that until the winnings are completely depleted, losses are coded as reductions in a gain, as if losing some of ‘their money’ doesn’t hurt as much as losing one’s own cash.” (p. 657)
Finally, risk aversion in the face of a previous loss can be overcome if the subject is offered the opportunity to break even (about 70% were risk-seeking in this scenario), especially with a modest outlay of money. The authors point to the proclivity for losing gamblers at the race track to bet on long shots at the end of the day. Although a risk-seeking bettor who is behind by $30 “could bet $30 on an even money favorite as a method of getting even, . . . [a] $2 bet on a 15-1 long shot offers a more attractive chance at breaking even because it does not risk losing significantly more money.” (p. 658)
The implications for the trader are obvious. First, it goes against the grain to pull the trigger after a losing trade and, by extension, after a string of losing trades (even when following a well-researched system) it requires an almost superhuman effort to take that next buy or sell signal. Second, winning traders get cocky because they no longer see themselves as playing with their own money. How many times have we read “move your stop to break even, then you’re playing with the house’s money.” The third result has two incarnations: the revenge trader who throws caution to the wind in his attempt to break even and the penny-wise and pound-foolish trader who is seduced by the siren call of out-of-the money, long shot options.
Monday, August 3, 2009
Kase, Trading with the Odds
Many of the indicators described in Cynthia A. Kase’s Trading with the Odds: Using the Power of Probability to Profit in the Futures Market (McGraw-Hill, 1996) are now available in the public domain for popular charting programs such as MetaStock, TradeStation, eSignal, Ninja Trader, CQG, and Aspen Graphics. The most intriguing indicator to my mind is what she calls the Kase Dev-Stop, and today I want to spend some time looking at its rationale.
Kase set out to improve on the standard volatility-based Chandelier trailing stop. This stop was normally calculated by multiplying a stock’s average true range (ATR) over a given period by some constant. For example, let’s say the 10-period ATR is 0.5 at the time of entry and the multiplier is 3. The trader is long XYZ with an entry at 20. The stock moves up to a closing high of 22 but then starts to pull back. If volatility remained constant during the course of the trade, the trader would be stopped out when the stock closed below 20.5 [22 – (0.5*3)]. If, however, volatility increased to 0.75, the trade would have more breathing room; in this case the stop would move down to 19.75.
One problem with this trailing stop method, Kase argues, is that “the level of noise is variable. This variability is not captured by an average, but by the standard deviation around the mean.” (p. 95) For the statistically challenged, Kase compares the height profiles of two different populations. The first group is made up of chorus line dancers; the second is an assortment of preschool children and basketball players. Let’s assume that we want to figure out how high a doorway would have to be so that 97.5% of each population could pass through without ducking. If the average height of the chorus line group is 5’7” and the standard deviation of the population is one inch, the door would have to be a little higher than two standard deviations above the mean—that is, a little higher than 5’9”—for 97.5% to pass through. In the second case the average height of the group is the same 5’7”, but the standard deviation is five inches. So the door would have to be 6’5” to accommodate the same 97.5%.
Another problem is that since volatility is skewed to the right (“bounded by zero on the downside and infinity on the upside”), “the distribution of range is not normal.” (p. 96) It too is skewed to the right. So instead of having “stop bands” placed at equidistant standard deviations, Kase suggests that there should be “a correction of about 10 percent on the second standard deviation and about 20 percent on the third standard deviation.” (p. 96)
To use the Dev-Stop Kase first draws a warning line, which “reflects the average two-bar reversal against the trend. The second, third, and fourth lines reflect one, two, and three standard deviation moves against the trend, corrected for skew.” (p. 97) Kase herself normally concentrates on the line that represents a three standard deviation retracement, though during highly volatile periods or if “in a profit-taking mode,” she will exit at the one standard deviation line.
The Aspen Graphics site, by the way, has a chart that displays Kase Dev Stops; it also describes possible uses for these stops. When I have some time I’ll test out Kase’s stop against the standard Chandelier stop to see whether it makes a positive difference to the bottom line. Readers can, of course, undertake this task themselves.
Kase set out to improve on the standard volatility-based Chandelier trailing stop. This stop was normally calculated by multiplying a stock’s average true range (ATR) over a given period by some constant. For example, let’s say the 10-period ATR is 0.5 at the time of entry and the multiplier is 3. The trader is long XYZ with an entry at 20. The stock moves up to a closing high of 22 but then starts to pull back. If volatility remained constant during the course of the trade, the trader would be stopped out when the stock closed below 20.5 [22 – (0.5*3)]. If, however, volatility increased to 0.75, the trade would have more breathing room; in this case the stop would move down to 19.75.
One problem with this trailing stop method, Kase argues, is that “the level of noise is variable. This variability is not captured by an average, but by the standard deviation around the mean.” (p. 95) For the statistically challenged, Kase compares the height profiles of two different populations. The first group is made up of chorus line dancers; the second is an assortment of preschool children and basketball players. Let’s assume that we want to figure out how high a doorway would have to be so that 97.5% of each population could pass through without ducking. If the average height of the chorus line group is 5’7” and the standard deviation of the population is one inch, the door would have to be a little higher than two standard deviations above the mean—that is, a little higher than 5’9”—for 97.5% to pass through. In the second case the average height of the group is the same 5’7”, but the standard deviation is five inches. So the door would have to be 6’5” to accommodate the same 97.5%.
Another problem is that since volatility is skewed to the right (“bounded by zero on the downside and infinity on the upside”), “the distribution of range is not normal.” (p. 96) It too is skewed to the right. So instead of having “stop bands” placed at equidistant standard deviations, Kase suggests that there should be “a correction of about 10 percent on the second standard deviation and about 20 percent on the third standard deviation.” (p. 96)
To use the Dev-Stop Kase first draws a warning line, which “reflects the average two-bar reversal against the trend. The second, third, and fourth lines reflect one, two, and three standard deviation moves against the trend, corrected for skew.” (p. 97) Kase herself normally concentrates on the line that represents a three standard deviation retracement, though during highly volatile periods or if “in a profit-taking mode,” she will exit at the one standard deviation line.
The Aspen Graphics site, by the way, has a chart that displays Kase Dev Stops; it also describes possible uses for these stops. When I have some time I’ll test out Kase’s stop against the standard Chandelier stop to see whether it makes a positive difference to the bottom line. Readers can, of course, undertake this task themselves.
Trading on Twitter
I realize that the second link isn’t exactly coming to you at twitter speed, but here’s a heads up on a new way traders are getting information and the algo guys are measuring sentiment.
The Wall Street Journal today, in its article “For Traders, Twitter Is One More Trading Tool,” looks at how traders in agricultural commodities are using Twitter.
StreamBase Systems is a software company that allows its users to analyze tweets in real time to inform trading decisions.
The Wall Street Journal today, in its article “For Traders, Twitter Is One More Trading Tool,” looks at how traders in agricultural commodities are using Twitter.
StreamBase Systems is a software company that allows its users to analyze tweets in real time to inform trading decisions.
Sunday, August 2, 2009
Winners and losers over the last 10, 2, and 1 years
We’ve finished another month in the markets, so perhaps it’s time to reflect on some long-term performance metrics for stocks, bonds, commodities, and currencies. This data is compliments of The Chart Store.
Over the past ten years, the best performer as measured by its annualized percent change was the Gold Bugs Index (HUI) at 18.32%. By asset class commodities were the clear winners. Energy commodities, with the notable exception of natural gas, gained 12-13% on the spot market. Industrial metals (copper, lead, nickel, and tin) were up between 11 and 14%. Gold was up 14%, platinum 13%, and silver 10%. Grains were all positive with gains between 7 and 10%. Among the softs sugar did best at 12%, cocoa next at 10%, coffee eked out an almost 2% gain, and cotton was fractionally negative.
Bonds, as measured by the Ryan Index, returned 7% at the long end of the curve and 3.3% for the 3-month Treasury. The strongest currency was the Canadian dollar at 3.41%, the weakest the Mexican peso at -3.32%.
And then there were the stock indexes. Measured by price only, not total return, the Dow Jones Industrial Average lost 1.49%, the S&P 500 lost 2.92%, and the Nasdaq Composite lost 2.84%. By cap size the winner was the S&P 600 SmallCap (+4.87%) and the S&P 400 MidCap (+4.42%).
Looking back two years, where would we have done best? We could have earned handsome returns by being very selective in commodities (soybeans, gold, cocoa, sugar) but would have suffered double-digit losses if we had invested in the industrial metals, in palladium, in natural gas, or in wheat. Bonds chugged along with returns between 2.65% and 8.99%, and the Japanese yen gained over 12%. All stock indexes with the exception of the Dow Jones Utilities Average fell double digits.
There are mighty few bright spots over a one-year time frame. Sugar is a stand out, up over 25%, the Japanese yen gained over 14%, and the U.S. Dollar Index was up almost 7%. Bonds held their own (though year to date the 30-year Treasury has cratered). Otherwise most of the stock indexes lost in the neighborhood of 20%, with the Nasdaq 100 best at -13% and the Dow Jones Transportation Average worst at -29%. The energy complex got hit hard, with one-year losses between 36% and 63%. The grains fell between 22% and 38%, about half of the industrial metals fell 30+%, the precious metals with the exception of gold (up 2%) fell between 22% and 33%, and the softs were all over the map (cocoa essentially flat, coffee and cotton down double digits, and sugar up 25%). A lot of currencies also took it on the chin—double-digit losses for the Australian dollar, the Brazilian real, the British pound, the Indian rupee, the Mexican peso, the South Korean won, and the Swedish kronor.
Over the past ten years, the best performer as measured by its annualized percent change was the Gold Bugs Index (HUI) at 18.32%. By asset class commodities were the clear winners. Energy commodities, with the notable exception of natural gas, gained 12-13% on the spot market. Industrial metals (copper, lead, nickel, and tin) were up between 11 and 14%. Gold was up 14%, platinum 13%, and silver 10%. Grains were all positive with gains between 7 and 10%. Among the softs sugar did best at 12%, cocoa next at 10%, coffee eked out an almost 2% gain, and cotton was fractionally negative.
Bonds, as measured by the Ryan Index, returned 7% at the long end of the curve and 3.3% for the 3-month Treasury. The strongest currency was the Canadian dollar at 3.41%, the weakest the Mexican peso at -3.32%.
And then there were the stock indexes. Measured by price only, not total return, the Dow Jones Industrial Average lost 1.49%, the S&P 500 lost 2.92%, and the Nasdaq Composite lost 2.84%. By cap size the winner was the S&P 600 SmallCap (+4.87%) and the S&P 400 MidCap (+4.42%).
Looking back two years, where would we have done best? We could have earned handsome returns by being very selective in commodities (soybeans, gold, cocoa, sugar) but would have suffered double-digit losses if we had invested in the industrial metals, in palladium, in natural gas, or in wheat. Bonds chugged along with returns between 2.65% and 8.99%, and the Japanese yen gained over 12%. All stock indexes with the exception of the Dow Jones Utilities Average fell double digits.
There are mighty few bright spots over a one-year time frame. Sugar is a stand out, up over 25%, the Japanese yen gained over 14%, and the U.S. Dollar Index was up almost 7%. Bonds held their own (though year to date the 30-year Treasury has cratered). Otherwise most of the stock indexes lost in the neighborhood of 20%, with the Nasdaq 100 best at -13% and the Dow Jones Transportation Average worst at -29%. The energy complex got hit hard, with one-year losses between 36% and 63%. The grains fell between 22% and 38%, about half of the industrial metals fell 30+%, the precious metals with the exception of gold (up 2%) fell between 22% and 33%, and the softs were all over the map (cocoa essentially flat, coffee and cotton down double digits, and sugar up 25%). A lot of currencies also took it on the chin—double-digit losses for the Australian dollar, the Brazilian real, the British pound, the Indian rupee, the Mexican peso, the South Korean won, and the Swedish kronor.
Saturday, August 1, 2009
Why trading's not like golf
Since I’m not pretentious enough to classify most of my ramblings as “bright ideas,” I’ll label them thought bubbles. Here’s the first.
Trading is often compared to golf (in many cases by those who are both traders and amateur golfers) and the super trader to the likes of Tiger Woods. Although trading is for the most part an individual endeavor, although I suppose one can liken the variability of market conditions to the variability of golf courses under different weather conditions, although both activities require extensive practice and mental conditioning, I nonetheless find the comparison wanting. Let me state for the record that I find golf boring and trading fascinating, so I’m biased. Here goes anyway.
The competitive golfer has a simple goal—to finish in fewer strokes than anyone else in the tournament. Each entrant pursues his goal independently of the rest of the field. It’s similar to taking a test and trying to get the highest grade in the class. The other golfers can’t interfere with his pursuit; golf isn’t croquet where a player gets points for hitting another player’s ball.
By contrast, most traders participate in a double auction system with innumerable interacting buyers and sellers. Every trader, from the one-lotter to someone on the prop desk of a major investment bank, affects every other. It may seem intuitive that big-time traders make more of an impact than the little guy—and probably 99.9% of the time they do, but think back to the analogy of the unstable pile of sand where adding a few more grains may barely matter or may make a huge difference (Mauboussin, 7/13/2009 blog). Even with inside information it’s impossible to know with certainty what the ramifications of the actions of any single trader or group of traders are, but we can say with certainty that in such a dynamic system there are ramifications. Intentionally or not, each trader interferes with the pursuits of other traders.
The golfer who thinks about what his competition is doing before taking his shot will probably psych himself out. On the other hand, the successful trader—whether systematic or discretionary—has to have a sense of what other traders have been doing in the immediate past, predict what they most likely will do in the near future, and estimate what the results of their actions will be. Otherwise, he can’t decide whether to dip his toe into the water, to become more aggressive, to go on the defensive, or simply not to play.
Subscribe to:
Posts (Atom)
Posts
