|
PureBytes Links
Trading Reference Links
|
Hi Herman,
Sorry to be late answering your post: I am visiting institutions in Europe.
I wish there is such a thing with a postential return of 100% a year. The
institutions at least have given up that idea and they sometimes spend
millions dollars a year in research. Maybe thet have discovered something
that the insividual investors has not. And don't forget that they operate
with very low transactions costs.
Backtesting over one year or so is edefinitely not enough. I would suggest
backtesting over a period of at least 5 years and make a backtest over 5 more
years. For stocks I suggest a blind test over the period 1974-1982.
I was having a drink with a friend of mine who runs a fund of fund with $500
Millions under management and he was telling me that , according to his
experience most of the technical systems crash and burn when effectively
traded. That's unfortunate but that's how it is.
As for backtesting with Metastock, in my opinion it's not a tool powerful
enough. For instance I don't think that you can get the daily Standard
deviation of your NAV.
In short, systems aiming at 100%, will very likely crash under stress i.e.
when a crisis occurs. But crisis happen all the time. If your goal is 15% you
will survive and maybe profit for it. Otherwise your exposure is too high and
, even if you were right , you might run into heavy trouble with unacceptable
drawdown. I saw recently some systems for sell and the results of the back
testing seemed good except that very often one would have had to add money
which meant that the system actually went bust.
A last thing: there is a lot of money to be made being able to produce 15-20%
returns inmmanaging money, most probably much more than an individual will
make with his own money making 50%.
However, it is worth keepresearchinbg. A system that works aiming at 100%
may be absolutely great aiming at 20% .
Good Luck.
Jean Jacques
www.alterama.com
|