How to do this is a much more difficult question to answer. Because you'll always run into the same problem. When the bear market kicks in, how will either algorithm know in order to keep AIM from buying what it thinks are cheap shares, only to discover the market is headed much lower.
Such is a problem not only for AIM, but most any "system" at all. No one has yet figured out how to get perfect reception through a crystal ball. I would suspect that even AI programs for all of their calculating intelligence are left stumped if all of the irrational humans decide to all jump off the cliff of selling like lemmings all at once!
I suppose one could argue for some sort of feedback loop particularly on the buy side, owing to the finite amount of cash reserve available that would act like a circuit breaker to slow down the buying process if under extreme duress. People do this manually to some extent by deferring sequential buys to monthly intervals. One way would be by resetting portfolio control to a prior level so that the next buy would be for the same dollar amount as the last, rather than being increased. This could come into effect if the price difference between the last buy and the current price exceeds a given safety margin (25%) - you could add a further test that this also kicks in if the cash reserve level is <= a given value, the idea being to maximize the cash reserve.
Just a thought.
Best,
AIMster
A
News 
Market Data 
Discover 
AI
