Steve Mayo Legend  Posts: 414 Joined: 10/11/2012 Location: Austin, TX  User Profile | The various covariates (factors) that impact system performance have differing relative impact. By that, I mean (and these are strictly fictional numbers), that the EF might account for 30% of the return while the strategy population might account for 30% and the account settings might explain the last 20%. It's a huge multiple-regression-statistics question, not unlike trying to figure out the relative impact of heart rate, kidney function, and arteriovenous intonation on changes in blood pressure. Ultimately, they all matter to the extent that even a small change in any one covariate is enough to give you problems over the longer term, whether it be a physical or financial stroke. [I do so love combining my two passions!] My point is that there may not be any one "best way" that works for everybody or in every situation. Sometimes, it's best to take a beta blocker to slow your heart rate, sometimes its better to take an ACE inhibitor to dilate your blood vessels; sometimes a diuretic is best. Trendline analysis might work better in certain markets while RTM approaches work better in others. Everybody will be different as to what side-effects (for investing, that would be level of risk!) they will endure. Repeat after me: there is no riskless investment; there is no drug without side-effects. :-) ECA is a game-changer because it gives us a totally new point of control, that being the EF and the periodic switching functionality. (It's sort of like we just discovered a new class of drugs for treating blood pressure!) And, unlike the confusing nuance of tweaking systems, lists and account settings, this new point of control is at a high level. It's like we've been trying to control blood pressure with only diet and exercise but have just now discovered how to directly dilate your blood vessels. In statistical terms, it is a major covariate with a high odds ratio, meaning it exerts a strong influence on performance (i.e., risk and return). But, it's still not the only covariate: system, list, settings, etc. still matter...a lot! There's a big fundamental difference switching and building. Our original concept was to switch BETWEEN PORTFOLIOS after building the optimum COLLECTION OF STRATEGIES. Although we developed an engine that can do both, simultaneously, Portfolio Wizard is primarily about strategy selection. The quantity and type of strategies used still exerts a strong influence on the results you get. Portfolio Wizard is giving us an incredible new way to directly and more-easily improve the performance of a portfolio. But, I think we still need the Portfolio Switcher to create a system that will truly react to material changes in the market. Switching is an even higher level of control than what is available with PW now. The EF that gives the best result when assembling a subset of strategies each month is likely not the same EF that will best detect a sudden change in market mode. The collection of strategies that work well in a choppy market is probably not the set that will work best in a rapidly falling market. Our latest efforts on using booleans in the EF and the move toward weekly intervals should get us much close to that goal, however. Meanwhile, PW is an incredible new tool...but I think the best way to use it is for improving the performance (and widening the scope/spectrum) of a DEFINED-PURPOSE portfolio, rather than trying to build a single BROAD SPECTRUM portfolio that is idea for every market condition. I think its good for deciding which drug in a specific class will work best for you, but perhaps not for choosing between different therapeutic categories. But, that alone is an incredibly powerful new capability we have never had before!!! The even better news is that we already have the engine in place to do true portfolio switching. It's coming soon! And, Ed just gave everybody a chance to get it for a bargain price! It will not replace human intelligence and perspiration just yet (which is why Elite Trader will be equally exciting), but ECA certainly moves the bar further in a giant step than anything I've seen since, well, the discovery of direct-acting vasodilators. [Edited by Steve Mayo on 8/4/2014 10:52 AM] | |
