Hello Philip,
I've done some research into SAFE settings and I've come to the following conclusions (which may change as I continue to research AIM and its myriad of possibilities).
1. Changing SAFE settings based on current market conditions is not such a good idea. AIM is meant to be a non-emotional mechanical system that derives all the information it needs to function from the stock price changes in your portfolio (after everything has been initially set up of course). It does not try to predict the next price, rather it reacts to it. Therefore if you change the SAFE settings based on how you or some analyst, talking head or cab driver feels about the current market, you are (1) introducing emotion into the system and (2) trying to predict the next price. For example, at the end of last year many thought the markets would rise by leaps and bounds for a very long time. If that was you, you might have been tempted to lower your buy resistance and raise your sell resistance. But that would have been the wrong action based on what happened from April to today. Of course if the specific company you hold in your portfolio changes its characteristics, then a SAFE setting change is warranted. Which leads directly into point number two...
2. My feeling is that different companies behave differently and there is not a one size fits all 10/10 SAFE setting. Robert Lichello did not have a smidgen of the resources and computing power we have today, so he did the best he could. I believe that if he had the tools of today (see automaticinvestor.com -- shameless plug ;-), he would have come up with alternative SAFE settings based on stock price characteristics. Somehow I don't buy the fact that RL came up with the ultimate SAFE settings for all stocks using a pencil and paper and limited access to stock price data. I think (and the historical testing I've done supports this) that stocks can be classified into general categories (based on the frequency and amplitude of its share price changes). Once we've determined the characteristics of each category, we can tune AIM to take advantage of this. Again that's why Automatic Investor (another shameless plug!) introduced models. That being said, mapping stocks into various categories is a bit more difficult than you might think. But I'm firmly convinced it will give better results. If you'd like to experiment with this send me a PM and I'll be happy to let you know what I've done so far. Once you've set SAFE, however, you should leave it alone unless the company reinvents itself and its stock price characteristics change such that it is put into a different category (for example Seagrams moved from a pure alcoholic beverage company to a large entertainment company after purchasing most of Universal Studios and other companies).
3. One of the things I've never been completely comfortable with when using the 10/10 SAFE settings (or any X/X setting for that matter) is that there is a built in tendency for stocks to rise over the long term (the good ones anyway) so I think the SAFE settings should reflect this. I would use an X/Y setting where X < Y. As to the exact values of X and Y, well that's the (literally) million dollar question.
4. I've also done some work with using technical indicators to dynamically change the SAFE settings and so far have been quite unsuccessful. Basically I've come to the conclusion that if I can find a good indicator (or set of indicators) that can be used to dependably adjust the SAFE settings, I might as well just use the indicators directly and forget about AIM. So far that hasn't happened.
However there is one area where I've had some success and expect much more in the coming year. And that's with using volume as well as price to determine AIM's recommendations. I've always believed that there is information inherent in the volume as well as the price. Therefore reacting to price only, effectively throws away useful information. I'm currently researching a way to use the volume. Of course this enhancement, if it ever comes to light, will have to go under the heading of AIM derivatives. So far I've been able to beat AIM (price only) results over 90% of the time when adding volume information and running historical analysis. But there's still much more work to be done here and I've only just begun. If anyone out there has any suggestions on this, please let me know.
Anyway, I hope this helps and welcome to the board.
regards,
mark. |
