Replies to post #27558 on A.I.M. Users Bulletin Board (AIMUSERS)

[Toofuzzy]

Hi Conrad

Other than the residual BUY that you find annoying (and I don't see as a problem) the only problem with AIM has to do with using an individual stock in an AIM account and having that stock go to zero. But then that is a "problem" for a "BUY and HOLDER" also.

The "problem" is eliminated by using funds in an AIM account.

Toofuzzy

[ls7550]

I think you might be looking too deeply into those charts Conrad.

Under AIM we can calculate the next buy and sell price levels given the Price, SAFE and minimum trade size values. There's a calculator for that purpose on one of Tom's web pages.

We call the range between the next buy and next sell prices the Hold Zone.

The charts therefore show how for consecutive buy trades how that hold zone narrows, and for consecutive sells how the hold zone widens.

The arrows just show the relative price level at which we enter into that hold zone after a buy or sell trade actually occurs.

At first when creating a new AIM account we start off with the arrow in the middle, approximately equal distant from the next buy and next sell prices. However subsequent trades in the same direction (e.g. a buy after a previous buy) has the arrow much closer to the next buy trade price level than that of the sell price.

As Vortex sets PC=SV then simply the arrows would be more central between the next buy and sell price levels in each and every case.

As a very crude example say that we start an AIM when the price is 100 and we calculate that the next buy and sell prices are 85 and 115 respectively. The arrow points to 100 and the vertical line spans the range from 85 to 115 (a range of 30 points). If then the stock price falls to 85 and we buy some stock then at that point the arrow points to 85. Calculating the next buy and sell price levels at that time will likely throw up a hold zone of something like 80, 109 (a range of 29). Such that the next buy price after that first buy is just 5 points away from the current 85 price level, whilst the next sell price is 24 points away.

Hope this helps rather than confuses you further.

Regards. Clive.

[Adam]

Hi Conrad, I want to explain my post better. All our methods are passive and buy on the way down. The returns from any of our models are very dependent on the security we test on.

For example if I take a security that goes down and never recovers this is the bane of all these methods. They do much worse than buy and hold, and the model that wins is the one that buys less on the way down.

On the other hand if the security recovers, then everything changes. Now they do much better than buy and hold, and the model that wins is the one that buys more on the way down.

So to test fairly we must use a realistic security. QQQ is fine but select a realist time frame where buy and hold will give something realistic like 5-20% over a year. And increase volatility just enough to yield some trades, not to dip to 50% then bounce back by 400%. That's not realistc.

I'm not saying the methods don't work, not at all, but if you're going to test and draw conclusions, test on realistic data.

Adam