Acrary, Thanks for this material - it is really wonderful stuff. You probably already have help, but if you would like someone to synthesize this material for you, I'll be willing to undertake that task.
everything here is true material. nevertheless, people should bear the follwing in mind: 1. it all depends on the quality=forecastability of the underlying systems. if you take all days of the sp future over a ten year period and you randomly select one hundred of them you have about a 3% chance that the modified sharpe ratio on your equity curve is above 1. the average sharpe ratio of alans models here is well below 1, thus be aware that you could be easiest fooled by randomness here. it finally is all about edge ... 2. whenever you correlate systems make sure you correlate their daily returns, not their net asset values, account balance or however you want to call it. this is essential otherwise corr will not answer the questions you ask. 3. make sure you treat return in the right way. an equity curve of 100 on t1, 50 on t2 and 100 on t3 did not yield 25% on avg per period ... using log yields helps with this. then the overall yield is 0%. peace
alan i wonder about one thing. i have never seen a sharpe of 1.6 with such a great sterling (return p.a./ mdd). from a 1.6 equity curve i would expect a sterling of 3, maybe 4 or 5. but 14? hmm. i hope i did not get anything too wrong here through my fast cross-reading your material ... peace
Outstanding thread so far, Alan, thanks. As an ex-buy side quant, I am enjoying it a lot. Sorry, but I am more than a little confused by that reply of yours from a few days ago. Maybe you can help? In the first paragraph, you talk about getting the systems to plug in (in a multi-system framework), not having to develop a thing, and maybe systems to buy. In the second paragraph, sounds like you are referring to a fairly long, hands-on, methodical, from-the-ground-up process of developing a system of one's own. Hmm... to me that suggests quite a different approach from the one above, would you agree? I suspect I am not the only one of your eager readers who would like for you to clarify what it is that we can look forward to, in addition to your working your way through showing us how to properly build a robust, optimal multi-system framework. Thanks again for doing this.
Yes, the two subjects are mutually exclusive. I wanted to show the benefits of building systems this way and then switching to the process of finding the models to plug in to achieve the results. My style of development requires a paradigm shift away from tactical trading (buy when x crosses y, etc.) to a strategic vision. Tactics are still there but only to supplement the strategic goals. So far it doesn't look like there's much interest in this material apart from professionals so I may take a break.
Please don't stop now. This is great stuff! Your descriptions of risk are very helpful. What software are you using for the MonteCarlo analysis? Your approach to identifying when your account should reach a new high (every 15 trades in your example) was very helpful. You described a very simple way to size your position. If you used this approach with your MonteCarlo testing would it appear to be too conservative and risk adverse? Would you change this basic position size approach if the market appeared to be more favorable to you? Have you found any way to match position size with market conditions instead of only using account size? For example, if last 4 trades were winners, increase the size of the next trade for example... Another example, if volatility of the market > x and trendstrength > y then increase size... Thanks again for the thread!
Alan: I always enjoy hearing from you on this subject. My approach is similar in many ways, probably because I have listened to your prior comments. I observe the following: You look to minimize risk first, then you focus is on consistency of returns. This is what professionals do. Amateurs in contrast look for historical tendency, and throw money at it hoping that risk will "take care of itself". It never does. On the system side, I have followed a rule that has helped me greatly and is probably of more interest to the retail folks. What I have done is to find repetitive market behaviors and incorporate both the behavior and "failure" into my systems. As you already know, if the repetition is frequent enough, and you have programmed correctly, you will then catch one or the other all the time. This works well IF both behaviors have significantly different results. The obvious example is a breakout after a test of a price level (like say a 50 period EMA) or a failure and move down to nearest support. personaly, if I were looking to make a dollar trading, after reading Alan's comments I would be mobilizing all my resources to try to learn to use a Monte Carlo Engine (I like @Risk by the way). Good luck guys, Steve
acrary, a big "Thank You" for sharing your process with us. Your discussion on the strategic design of systems is one of a kind. Please don't stop! I'm sure there are many like me that are quietly absorbing your posts and anxiously awaiting the next one. I have a question for you in regard to your correlation and weighted models work. Does each model added trade one instrument such as ES? Or does the model trade an edge across several instruments? What if you have a trading model that trades stocks in the Nasdaq? Would it be appropriate to keep that as one model? Or possibly break the stocks into their perspective sectors and have numerous sector models where you could use your correlation and weighted designs to add each sector model only if improvement is made to overall return and risk? Of course, by breaking up the 1 model into numerous sector models...that would enable sector tweaking of the original edge. Not sure if this would be the way to go or not. I'm currently working on what I'm asking but would be very interested in your insights/experience. Thanks again!
Acrary, dont stop here. Altough i dont have that many robust models yet, i find this stuff very interesting. I hope i have 3 good models by the end of the year, so this model weighting would become extremly valuable for me. And thank you so much for sharing!