Re: [amibroker] Do all trading systems stop working? - Howard Bandy's book, AmiBroker Email List Archive

Hi all,

How do you define "failure"?

Is the nature of it's failure important? (ie it might fail
quickly, or it might fail slowly).

From my experience, it is actually very hard to quantitavely
determine if and when a system is actually failing. I still
don't know and I've been trading for over 10 years, 6
or so as my sole source of income.

A lot of reading that I've done seems to indicate that a
lot of strategies get arbitraged out of the market over time,
but do not totally disappear. One example was a study of
some common pattern based TA in the US futures markets
- Head and Shoulders, triangles etc. It seemed to indicate
that returns from these patterns have been diminishing
gradually and steadily for the last 20 or so years.

This makes sense. I think that there is an underlying
assumption in TA that statistics for markets don't change.
I think this is false. Markets are a game ( in the
mathematical sense.) You aren't measuring an absolute
attribute in nature, you are trying to find a strategy
that beats your opponent(s).

I like to think of it this way. Can you sit down and
gather statistics of all the chess games ever played
and use those statistics to become a successful
player ( for a given value of success)? Maybe,
maybe not.

When Deep Blue played Kasparov,
Deep Blue "analysed" all the games that
Kasparov played. Kasparov complained that he
did not get a chance to look at Deep Blue's
"learning" games. So statistics definitely help,
but I don't know if it's the be-all end-all. It's
also important to know that Kasparov felt a lot
of pressure, and (as far as we know) Deep Blue
did not.

To be successful in trading, I think you need a
strategic overlay to the statistical insights offered
by TA. I also think that the objective nature of
TA takes off a lot of the pressure in diffucult times.
It's during those difficult times that you will be
making decisions which have the most
potential to "make or break" your account.

Robert Z

From: samu_trading <samu_trading@xxxxxxxxx>
To: amibroker@xxxxxxxxxxxxxxx
Sent: Sunday, 31 May, 2009 3:44:19 PM
Subject: [amibroker] Do all trading systems stop working? - Howard Bandy's book

All,

In his really good book Quantitative Trading Systems, Howard states that all trading systems will stop working forever at some point (because the inefficiency in the market they exploit will be killed by everybody jumping on board).

On the other hand you have momentum / ROC based systems working forever now, same for trend following MA crossover systems like The one propagated by Mebane Faber. Momentum and MA rossover trendfollowing does seem to work "forever".

Any comments from the gurus here?

Thanks, Samantha