Obviously I realise that one OPTIMIZATION RUN for all stocks in the world would not give exactly, an average CER= 50/50 but if you are right and we do the optimisation experiment 1000 times the average of the optimised CERS must be 50.50. . if you are right, and if the result is not 50/50 on the average that would mean that a starting CER of 50/50 is not generally an optimum value.
Not sure if this isn't getting into the "how many angels can fit on the head of a pin" type of debate from the Middle Ages, in the sense that one can determine the optimal ratio using past data as if you've invested back in the past. The problem is, as I see it, that that doesn't really count for all that much for what ratio to use if you're starting a new AIM program today. Whilst the v-Wave may be an adequate measure for the relative few of us who've managed to find our way here, I think Clive's point, for the ordinary "John Q. Public" investor is that 50/50 might be a reasonable starting point. This was, after all, Lichello's earliest opinion when he found the "2X" marking on the license plate: the odds are 50/50 that the stock will go up or down from any starting point in time, especially if you don't bother to look at a current trend of recent price activity.
In the world where we all would have perfect crystal balls we could know the optimum in advance. But one test, or even 1,000 will only show you optimum numbers for the range of time and the range of stocks selected. I suppose to get a realistic number would be to number-crunch the 1,000 or so tests together, repeating all 1,000 for each trading day and see what the average numbers come out to be over a protracted period of time. Though again, it seems to me to be more an item of curiosity rather than truly practical information I could use today.
Or am I missing something here?
Best,
AIMster
