Replies to post #3787 on A.I.M. Users Bulletin Board (AIMUSERS)

[Rien]

Hi Tim,

I think that's the gist of the excercise - connecting the AGD and RSD through some characteristics,

This is the real problem. Something for which I don't have an answer. We can test all the stuff we want but in the end we have to make a decision on which stock to AIM and with what parameter settings.
I am afraid that the stocks that produce good returns might not have much in common, at least superficially.
For example, say a certain stock produces 40% returns, then it might well be that a minute change in the chart can change this result by 10% or more.
This was in fact why I want to be able to produce AGD without any random functions in it. And also why:

I'd call this an intuitive understanding of the process; that's a subjective goal.

I think this is very important. I do not expect that I will be able to come up a "proof" that AIM on a certain stock (i.e. with defined characteristics) will always produce X% results.
But I will be able to enhance my subjective understanding, which should allow me to make a better guess at which stocks to AIM and with which parameters.
Of course it is also possible that the result will be "just throw darts". There are indicators that this outcome could be rather likely.




Best,
Rien.

[OldAIMGuy]

Hi Tim, Years ago I had a cute little DOS program called "BLACK MONDAY". It was written some time after the '87 "crash." It had a random number generator in it along with possibly some higher level trend development math included.

The idea was to play out the game. It gave you a long list of stocks with real names that you would recognize like GE and IBM, etc. You could allocate your money any way you wanted in the game. Lots of stocks or just one. You could also trade it as often as you wanted. It kept track of your profits and losses. Sometimes after a random number of price updates (you never knew when Black Monday was going to come!) prices would crash terribly. Other times the game would end with no severe crashes or a crash and a recovery. You never knew. It was like an electronic Russian Roulette!

You might win big early by trading successfully. Then when the crash came, you might still be profitable. If you hadn't made quite a bit of money in advance of the crash, then you most certainly were going to lose. Since CASH was a real option in the game, you could just stay in cash and always win at least to the extent of the interest rate that it applied.

It was wonderfully fun and instructive.

Well, I decided to see how AIM would do with the game. Using a pad of paper and a calculator, I played the game while using AIM as the guiding light. After dozens of tries, I found that AIM almost NEVER LOST. When it did lose, it lost minimal amounts compared to the cowboy day trader style.

AIM was so good at retaining its gains that I think I frustrated the game! Now, one more important point. When playing the game as a trader and maintaining some discipline, one usually out performed AIM through all the rising periods. By being fully invested all was risked and all was rewarded. Win big or lose big. AIM's winnings were moderated but the few number of loss games were so tiny that overall AIM did quite well.

With the market the way it's been, probably someone should dust off that old BASIC compilation and make it a fancy Windows program. It would probably sell well right now. One feature it lacked was that it wouldn't let multiple players run it at the same time to determine a "winner." If I played it once with AIM, and then played it again as a trader, it was never the same because of the random price histories and time frame. It would have been fun to be able to run multiple strategies simultaneously with the same exact histories and time frame (much like we can now do with current software and spreadsheets).

Actually, since the program couldn't be "tweaked" for optimum performance, it was much more like real life. We never know what the next random number is going to be. Tweaking is useful as a retrospective activity. I think it would have been next to useless, however, with this game. What the game needed to be a consistent winner was a solid and logical approach to capturing profits. I was delighted to see that AIM, in its most basic format, did exactly that.

Sometimes it's now how much one wins, but how well one avoids losses that determines success.

Best regards, Tom