[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Trading systems....the path to the holly grail...



PureBytes Links

Trading Reference Links


Hy,
 
First of all my english isn't that good but i 
will try to explain myself.
 
Good tradingsystems like holly grails do excist. 
I work with one. But....
More and more i am convinsced that they are VERRY 
VERRY hard to make on E.O.D. base.  I work with 2 systems. One on a 5 min. 
base and one on a 30 min. base. and they work great(understeatement). The 
tradingsystem on an 5 min.base gives me for this year 1.500% profit(yes you read 
well thousandfivehundred) and the the tradingsystem on a 30 min. base gives me 
for this year 1.200% profit (thousandtwohundred). I can asure you that formulas 
for those systems are very simpel. I believe in simpel things. Now i will come 
to the point. If you make a system on a E.O.D base and lets say you build a 
system and you test it on hystory data. You have to thake enough of data. Lets 
say 10 years. One day is one piece of data. That makes that you have in one year 
220 pieces X 10 years..in total 2.200 pieces of data to work with. The first 
thing that you see if you look to that chart of the last 10 years that you 
will have a verry bullish chart (that counts for 99% of the indexes 
and stocks). So if you will do an optimization it will make all your profit on 
the bull rallys. But it will make losses on consolidation or bearish rallys 
because you dont have enough information in those charts about bearishness. If 
you will look to your test results it will make profit but the draw downs in 
your equityplots are horrybull. Also because you only have 2.200 pieces of data 
you will have to few trades to have a reliabel system. NOW...If you make a 
system on a 5 min. base on for ex. XETRADAX you have 12 pieces of 
information in one hour X 11trading hours a day is allready 1.320 pieces of 
information in one day X 1 one year that makes = 290.400!! pieces of 
information. If you do your backtesting on a chart of the year 2000 it will 
give you bullishness and bearishness because that is verry important!! that you 
have all kind of market situations. A good system is one that you want drown in 
trades. 3-5 trades a day is good. The equityplot is NEVER showing a 
drawdown bigger than 5% of the equity that my system works on. That 
counts for my backtest aswell for the the compleet period this year that I realy 
work with it and thake all the positions for real.
I'm sorry that like most of the people who looked 
and searched for the holly grail for years and found it I will keep my system 
for myself....but I can give you a little hint to build one. You need on a 
intradaybased system only 2 things...and that is a good trendindicator and a 
good oscillator. The oscillator helps to trade ONLY with the trend. A 
trendindicator who can do the job is for example an MACD but....MACD > 0 is 
bullish and MACD < 0 is bearish...if you would interpret the MACD in a normal 
way it would give to many signals. If your trendindicator shows bullish than you 
thake with your oscillator ONLY the long trades. Bearish trend is with 
oscillator ONLY short trades. As oscillator on 5 min base is an Stoch. 9-5-3 a 
step in the good direction.....The 2 indicators here just placed are 
not the ones that I use but it could help you on your path to the holly 
grail...to understant that the answer lays in a simpel aproach....If you find a 
trendindicator that gives good backtesting...and you find an oscillator that 
gives good results....and you combine the both of them...acsidents are verry 
rear to happen. But again this is all very dificult to inplement on a E.O.D 
base
 
Again sorry for my english ...but I hope this 
makes some sense.
greetings igor.
 
<A 
href="mailto:igor.devisscher@xxxxxxxxxx";>igor.devisscher@xxxxxxxxxx