did you use randomly generated price data or did you use actual historical market data? if you did the latter, i believe this is similar to what Van Tharp did on commodities markets (pls correct if i am wrong). it has been well recognized and documented that some commodities markets exhibit significant positive serial correlation, i.e. trend on a longer term basis. to take advantage of it, one must stay with the trend for as long as possible, i.e. so that the tiny day-to-day positive serial correlations cumulatively add up and compound into a significant end point total profit. in this case, the exit strategy itself does not have an edge, the stay-with-the-trend exit strategy is only a means to extract the edge from the positive serial correlation phenomenon.
i look at it this way: you trade your capital because trading it is the optimal risk/reward with respect to other opportunities for employing your capital - if you miss a trade, the opportunity cost is as much of a loss as an actual loss on a losing trade, i.e. the opportunity cost is as real and quantifiable as an actual loss - hence missing a trade means losing, indirectly, and thus the only reason to miss a trade should be a technical problem, human error, etc. which Eckhardt brings up in his comments.
Hi billp. For this test I used a 20 day ATR and a 10 ATR stop from the trigger day´s close. Since it is so wide, it can sit there for years, with the market going nowhere. Note that this was just an exercise, I would never do this myself, not without some minimal indication of where the market might be headed.
Hi fader. Yes, I used actual historical market data, pretty much the same way Van Tharp talked about in his book. However, I respectfully disagree, in the sense that I believe the exit actually does provide an edge. Like you said, it is staying with the trend for the longest time possible that produces profits, and this is provided exactly by the exit, which gradually captures profits while giving room for noise. However, I wholeheartedly agree that exits are not enough to produce repeatable results. My point is that your exact entry is not so important be profitable. IMHO, the entry is important to produce superior and less risky returns. Once again, this is just my personal opinion.
wow, I got impressive results too trading a random entry. I also randomized the position sizing ( within a 1% range) and got even better results.
Tireg, Thanks for starting the thread and giving the additional references. The two most surprizing things to me were the equity chart and paragraph 5 of your intial quote. The equity chart seems to compare the compounded result (I am assuming that some equity is removed for some things not mentioned) and the value of the S&P in some way. It looks like apples and oranges to me. I do not keep some kinds of basic historical records. If anyone has an equities chart of what the S&P results would be if a sum were invested the same year and got a compountded result equal to the gain or loss of the S&P annually I would like to see it. I have comparisons of the " 3 fat cats" for specific ten year periods and I have a mutual fund comparison done three ways as well. The equity curve shown does not have a apple to apple type comparison though. It looks like the annual average rate of return is not too swift at any rate. Paragraph five is a puzzle to me. here it is for convenience: An important feature of our approach is that we work almost exclusively with price, past and current. One reason for this is that to make any progress in the early stages of quantitative investigation you usually have to reduce the relevant factors to one or two crucial variables. Price is definitely the variable traders live and die by, so it is the obvious candidate for investigation. The other reason is that in a system thatâs making good use of price information, it is very difficult to add other information without degradation. Pure price systems are close enough to the North Pole that any departure tends to bring you farther south. I have been an explorer and under contract with Parks Canada I've been pretty far North so I find the analogy to be amusing. My job was to explore for places to open to the public because of their environmental, scenic and recreational value. In the early stages of anything, a person has to keep his eyes pealed and have an open mind for any possibility, in my opinion. Eckhardt didn't he tells us. I believe he was focussed on his training and applying it to markets. Traders do not focus on markets in the way eckhardt did, either. Mostly because they have no related training nor do they approach trading from a general and open- minded point of view. Most people do come to the place Eckhardt occupies with respect to the variables of the market. His paragraph, when read slowly sees him slipping away from the reality of markets as he says: "One reason for this is that to make any progress in the early stages of quantitative investigation you usually have to reduce the relevant factors to one or two crucial variables. Shutting down on qualitative analysis is a big step. Going from two crucial variables to just one, is another big step. Eckhardt took the steps and was saddled with the results he has to show for it. What if he could have forged along by doing something more than that which traders live and die by? I think he would have gotten a different paradigm than the one he has arrived at. He jsut gets a fee and a percent of something that works at a certain quantitative level. Anyone who is building an approach or method to "understand" the markets and then "set up a business" to run their or OPM, has an obligation to work with all the direct and independant variables (with respect to time usually) that, in fact, define the markets. It probably is true that diferent variables are more or less important. It appears that Eckhardt chose to interweave trading and market operation and only base this merged thing on price. He is also proceeding to the limiting case where he is "biggering and biggering" (See the Lorax by Dr Seuss) in terms of computer power to go deeper and deeper mathematically in the hole he is digging at the north pole. One thing that emerges from this is to consider what the parallel paths would look like if these paths were based upon different types of formal or informal training. Well that is easy to do and it is in the literature all over the place. How do these approaches size up and how are they ranked by people who take the trouble to do it? My conclusion is that it is an interdsciplinary opportunity. How can an interdisciplinary opportunity be carried out to get to a pracmatic solution? I think it takes a lot of looking around and with an open mind. It looks like to me that people who are improving what they do, spend their time getting tools, extensions of themsleves, built for only about three or four purposes: monitoring, analysis, decision making and taking timely action. Its probably time, on ET, to begin to present what these tools can be; why they came to be; what their true purposes are and how they produce improved results. You can see the four purposes of tools with Eckhardt; you can also see how the invented tools of Eckhardt have the many sides mentioned above, all subject to the constraint he chose to not do qual nor consider more than one variable. Whats next?
From speaking with many people, and much reading about risk, Kelly ratio cuts it too close for risk of ruin. Some people use a variation of optimal-f, which is smaller than Kelley. Ralph Vince's book Portfolio Management Formulas explains in detail this concept. As far as knowing the distribution of odds, here's the thing: you may know that over the next 10 trades, 5 of them are likely to be winners, but you don't know in which order they will be. Would you be willing to bet full Kelly on the next few trades? That is almost guarantee for some deep drawdowns. Optimal f on the other hand, is supposed to be trade order independent. But sure, you might say. I'll just move the stop closer. But a largely unknown idea for you that I've just learned about is that the closer the stops are, they more they change the probability of the outcome - on an individual trade, as well as on a portfolio level, in a negative manner. I refer you to this study, by Robert Macrae of Arcus Investments: in PDF
Hey Grob, thanks for your input. The equity chart is amazing when you look at it on the surface level. Not many can say they can provide such a linear and near-optimized ride for their clients. Digging deeper, we can look at his annualized return and maximum drawdown; 26.24%, and 29.08% respectively. Not awe-inspiring numbers by themselves, but these numbers would indicate that sheer consistency of which Eckhardt's run the money is what has achieved this smooth curve. We can dig a little deeper now and look at how this curve was actually achieved. If we look at the yearly returns, we will in fact notice that his first decade of trading brought him most of this. Having a blast-off start and kind of petering down would be how I would describe it. Personally I think that the cumulative return chart can be misleading, since it starts off with a small number, 0%, and any increasing % is indeed compounded. As such, emphasis is put on high early returns to achieve a large percent gain. Looking at recent performance, his returns have been hit and miss... with some years failing to beat even risk-free rate. I do not know if this is due to a change in risk tolerance or change in strategy or possibly both. Would I invest with him today? Without understanding the reason for such performance dip, probably not. The little red graph there, on the bottom, I assume is the S&P 500 Index returns. I would agree that comparing the two are like apples to oranges, but investors like to see some sort of 'relative return' comparison. I would argue that as a hedge fund, your responsibility is for 'ABSOLUTE RETURN', regardless of market conditions. I feel that many fund managers today have forgotten the 'hedge' part of fund. Many funds are just leveraged directional bets, without proper compensation for risk. With the insane amount of hedge funds out there, it is a good thing that survivorship is a factor. On the other hand, unfortunately a side effect of this type of behavior, investors and institutions, FOF, are more finnicky than ever and the propensity of them pulling their money is much greater, particularly at the most inopportune moment - re: drawdown. As a result, many funds that might have otherwise been successful are forced to close prematurely. But I digress. Comparing yourself to any index would defeat this purpose of ABSOLUTE RETURN. However, to compare similar funds, many often compare fund performance vs the strategy-index performance, such as a long-short hedgefund and a long-short index. As for paragraph 5: I agree with your thoughts here. They are better articulated than my earlier posting about the academic view of 'right and wrong' and 'fully accepting or rejecting' a concept. In the scientific and academic world, one has to isolate variables for study in order to create the 'right' answer, but in the financial world, such blind focus can often lead one to miss opportunities. You summed it up quite nicely: I think you bring up a good point about a scenario with formal and informal training. We can use Eckhardt as an example of one who has had formal training. Hypothetically, an informally trained person would focus on what works or what is practical instead of what things should be. They would likely see the market in more open eyes and how it actually is, than to skew or bias their perception by only focusing on one aspect. At the same time, though, we must look at the cons: namely that an informally trained person might be more likely 'fooled by randomness' because they may draw spurrious assumptions and conclusions from their results or the results of others. The lack of formal training may also lead to many inconsistent results, since the person may or may not understand what is going on in the big picture. As a result, it is important to understand the tools one uses as well as the results of the outcomes. Keeping an open mind certainly helps. Which brings us to your final thoughts: I commend you for promoting this ongoing exploration and discussion with new challenges. Not many here have done so. I am still in the early exploration/learning stages, as my views on these topics must reflect, but I would be glad to discuss how I am going about things. Hopefully others will contribute meaningfully as well. I would also turn the question back to you to see your thoughts on the matter. I think that would be best reserved for a new thread though. I wanted to keep this particular thread focussed on Eckhardt's reasoning and strategies.
Fact: NOBODY on ET is even close to being in the same league as Eckhardt. Yet many of you have dismissed/criticized his thoughts and performance as if you know better. You people never cease to amaze me! LOL