Clive, I am beginning to feel very guilty for making you go so much trouble. . .
(Sorry)^3,141
After reading this I feel sorry I asked. . . Now I feel obligated to make at least an effort to "get it"
OK, I understand that the LST is supposed to be used to calculate the trade sizes. . the that worked went over my head. In my example cases I used an arbitrary set of trades to sell and to buy, simply the way it "looked" about right as a rising percentage of the remaining "object function". . Reserve Residue for Buying side and the Equity Residu for the selling side. Before trying to relate the Trading Steps to the LST there remains a missing element in my understanding, so before proceeding I would like to "grasp" the significance of the difference between calculating the progressive step sizes via the LST-Difference function and the way I chose arbitrarily a progressive trade function. . .essentially such an arbitrary progressive function can easily look very much like to an exponential form. The only difference then is that my progressive function is note pure exponential, but within in the narrow range the trade sized would be very similar. I think I understand that with the LST we would essentially create a pure exponential Trade Ladder. . .right?
Why should an arbitrary progressive Trade Function be less effective that one that is made up by using the LST Function?
Would somehow the progressive Trade Function from an LST procedure be superior to the arbitrary progressive Trade Function? I would think not, or else I fail to see why. . .the LST function also has no input from the stock process in “between” the Action Points and therefore it can not predict a better a trade function. . . shoot me down if I am wrong.
Considering the fact that for an investment we would be ready to sometimes skip a trade, and uberhaupt we would be executing trade values based on nicely rounded-off number of shares. If the trade advice would be 7853 shares I would trade 7900 or 8000 or even 7000 units, depending on what I would be thinking at the moment. . .Any trade advice system would be a rough indicator anyway!
If you agree with me that any progressive function that would look about right will do just as well IF it satisfies the trade sizes I approximately want to implement. The only thing that remains is the choosing a system that is most easily incorporated in the AIM program so that it supersedes the Vortex AIM Algorithm.
Right?
In the meantime I will study your answer in more detail and try to grasp the details of it.
Regards,
