A.I.M. Users Bulletin Board

[Conrad]

Hi TooFuzzy, you are indeed making it to fuzzy

"Your formula makes it look like you are dividing but i am pretty sure you are not. I will write it out slightly differently but it is probably the same.

>>>>>AIM Trade = /(PC-SV)/ - a*SV <<<< "

This IS exactly the same as I put it with /()/ being the absolute value. The only thing to choose for a particular AIM-portfolio is a choice of a. The AIMer involved in a comparative test should choose it. . .the reason "why" is not relevant to me. The only trades (in the runs already carried out) that are executed are => than 3500 AND, if applicable the 5% Hold Zones comes into play if a trade under the 5% price change triggers a trade > 3500. . .in that case the trade would be executed only at price change => 5%.

For the First Reference Trade, in the test-case, +/-,

% Price Change = % SV Change

and that is relevant for defining the Reference Trades so that AIM and Vortex start off identically for a rise or fall in prices. How in an AIM portfolio the second and following trades +/- are calculated is of no concern to me.

The only interesting thing is to find out how, under identical Min. Trades and identical Holding Zones and identical "No Negative Reserve" rules the trades differ for a particular Data Set.

Therefore, how you or anybody else would define the value of a in general makes no different at all as long as it is defined for a particular case. Since the price development for a real stock is unknown the choice for the parameters is not relevant. . .one only can speculate what type of price development is expected and then pick the parameters.

In our case the comparative testing the choice of the parameters is completely arbitrary and I will accept any value of a one cares to select and how it is selected is not important to me.

The other things you wrote with an A (instead of an a?) are quite fuzzy and are i.m.o. probably not relevant for the testing we are doing.

With this in mind any AIMer that wants to do a Comparative Test (not a competition!) may provide any Data Set and any starting condition and run the AIM with that data. The only thing I ask is to limit the data to 36 points or so, as I have to enter the dates and the prices manually in my Excel template, unless the figures are provided on an Excel File and send to my e-address. Alternatively I could download an AIM Excel Template and copy the price data from that.

To make it simpler yet, the AIMer may give me the initial conditions and the First Reference Trades from a starting price. The first trades need not be at equal percentage price change. For example you may provide

SAFES(Holding Zones) %. . (arbitrary x% and y* of the starting price)
Starting price
First Sell (value) at X% price rise. These are not necessarily the Holding Zone percentages but could be,
First Buy (value) at Y% price drop
Min. Trade: you pick it
Condition: No negative cash allowed.

How AIM prices are calculated is of no concern to me as long as you do not change the parameters in between.

The analysis of the results is primarily focused on the Trading Response. . .step by step, to the test data. The yield is of course also important for judging the effect of the trading but this will become only useful for deciding how one is to change the parameter set for a particular fund. . .I think that will only become clear after doing more studies. I do not expect to find an definitive answer to that in a short time. If someone does know how to provide that definitive answer I would sure like to hear about it!

Regards,