Q.
Tom,
Since I know you are a big AIM user, I wanted to tell you and the others that I have developed a new AIM program which I feel has many neat and exciting features. The program, which was written in Microsoft Excel, employs a backpropogation learning algorithm (if you don't know what that is, don't worry) that samples historical data from an equity and then calculates the best Buy Safe %, Sell Safe %, and Portfolio Control Factor. This enables you to modify AIM slightly to maximize your return by adjusting the program to the
"personality" of the equity. In addition, you can also input values (stock price, shares, Portfolio Control, etc) and have the program tell you if you should Buy/Sell/Hold, Market Order, and Shares traded. Also, the Buy Threshold and Sell Threshold are computed so you know exactly what price the equity must fall or rise to, in order to trigger a buy or sell signal.
I will finish up the final touches on the program within the next week. Also, once I get access to SI, I will post a message there as well for those who might be interest in the software.
Aiminvest@aol.com
P.S.-- Portfolio Control Factor is the term I use for the percentage of Market Order that is added to your current PC when making a buy. For instance, Lichello says to take 50% of your market order and add that to PC. I have found in my research, that 50% is not the ideal. In fact, in most cases, 54-58% generates better results. My software calculates the best PC Factor, as well as the best Buy Safe and Sell Safe.
A.
Dear AIMinvest,
Sounds like you've put in a goodly amount of time on the AIM template for Excel. My first AIM template was a Lotus 123 piece (circa 1988) that worked quite well. However, it carried all the usual Lotus bloatware with it and was a bit slow.
The Split SAFE is a great single improvement on the original AIM. I've always been surprised that Mr. L didn't come up with the idea. We've been using that since 1989 or 1990. I had attempted to get in touch with Mr. L at that time to get his thoughts on the subject, but received no reply. I wrote to him via Signet Books. I'm not even sure if he's still alive. I keep hoping one day he'll show up on SI or in my Email in-box!
Over the last few years, several people on Prodigy and SI have made attempts at adding some artificial intelligence to the algorithm and at least two were following the same path of an altered PC addition. I don't know if they ever completed the work, however. It sounds like you've carried it to a higher threshold. Congratulations!
When Mr. L conceived of AIM, the market had been really stinky for nearly a decade. I really don't think he planned to see a market like the 1980s or 1990s. AIM "By the Book" didn't account for a steady or unsteady rise over an extended period of time. Since he never capped the cash reserve in any way, AIM's overall liquidity was absurd after an extended bull phase. That lack of an upper limit on Cash Reserve sent many potential AIMers off onto other paths.
At the AIM Users Group web pages, you've probably read about the artificial cash cap I've been using on AIM and what to do when that ceiling is reached. This is another great single improvement. In my opinion, an AI feature that would maximize total return would address the maximum "draw down" of the stock over an extended period of time and use this within the calculations. I "ballpark" such values and compensate my portfolios accordingly. With ten years experience of AIMing, it seems natural to me. AIM Newbies don't have this luxury of experience.
AIM's only nasty habit is that it tells people "I told you so!" when they try to cheat AIM! When the price is changing and AIM's calling for a BUY or SELL, they decide that they can wait another day and see if they can get an extra buck out of the trade. Then the market turns and they miss the trade for now. It's my opinion that once a business plan is adopted, it should be followed. There's room for modification, but only at "annual meetings" and not daily! Micromanagement doesn't compensate for missed opportunities. We have lots of "shoulda, woulda, coulda" reports on the Bulletin Board regarding this.
AIM's biggest benefits are risk management, asset allocation and market timing. An investor must have already sensed a need for one or all of those before he'll allow himself to be "baptized" in AIM's ways. It performs all quite well right out of the book. I think the improvements we've been making to the algorithm collectively are making it better. Mis-application is still the biggest single problem new AIMers have. As Mr. L likes to remind us, AIM still won't make a bad company good!
Please let me know when you have the program up and running. I'll be happy to list it as a resource at the website. Bill Reideman has a working Excel AIM spreadsheet available, but I don't believe he has a PC Adjust feature in it. Another fellow just introduced his DOS based "stand alone" version and many of us use the Newport Windows "stand alone" as well. I'll get the names and email addresses of some of the others if you would like to contact them. Thanks for the note, I look forward to seeing you on the SI threads.
Best regards, Tom in WI |
