Some thoughts in no partiucular order...

1- PCA (AIM) works best on a ranging market. You have to adjust
sensitivity for next buy/sell to certain characteristics of the stocks
usual movement. For instance if you set parameters to trigger buys/sells
based on 20% excursions on a stock that only moves 5%, you'll never see
a trade; similarly if you assume 5% excursions and the stock typically
moves 25%, AIM will have you chasing the stock with sequential trades
which will raise your trading costs. The latter example is probably
preferred if you feel comfortable with 2- below.

2- Performance can be improved somewhat (but there's no garantees) if
you time your buys & sells using TA. AIM will tell how much when you
finally execute the order.

3- AIM doesn't use an averaging process. The algorithm is quite specific
and will suggest you to buy or sell only when price has moved the right
amount from the last sell or buy. It will also adjust how much you buy
or sell as a function of the price change. An averaging scheme will have
you buy independent of price movement. Lichello's book includes such a
scheme which has a little more intelligence built into it for those with
limited initial resources whi are just starting out.

4- Fundamentals need to be right for whatever you trade. That's the only
way to minimize the likelihood of something collapsing and not
recovering. Note that I said minimize losses... there are no garantees.
If the fundamentals change, it's best to ride a different horse.

5- There are no garantees. You can fiddle with AIM parameters till the
cows come home and the market will still do whatever it wants.

6- Back testing only tells you what worked over the backtest period and
may not have anything to do with what you see going forward.

7- As a guide, choosing stocks with good fundamentals which also have
very high or very low Beta will usually provide the best candidates as
they exhibit more price movement than the S&P 500. Also, if you can find
the data, a positive Alpha will improve results as a positive Alpha
indicates a security which is generally increasing in price.

Just a few thoughts for using AIM.

Bruce
Santa Fe, NM

----- Original Message -----
From: "rvlv" <rvlv@...>
To: <positioncostaveraging@yahoogroups.com>
Sent: Friday, June 05, 2009 4:36 AM
Subject: [Position Cost Averaging] Weak zones of pca performance


> hi
>
> PCA is configured based on certain assumptions.
>
> 1 It assumes that the scrip you have chosen falls from 10 to 4 and
> springs up again over a time period.
> 2.you use a averaging process as price goes down.
>
> How do I protect my capital when the stock I chose falls from 10 to 4
> and never gets up again?
> What kind of precautions I need to observe while choosing my stocks?
>
> pca does not choose scrips.
> I have to choose them.
> If I had chosen a scrip that has no fitting characteristics to pca
> then?
>
> Hope I can get some clear answers to these questions.
> cheers
> rvlv
>
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>
>