Trading System Question

J

JennaQ

Hey Guys,

I've been looking at this forum for a while now and I'm pretty impressed with the depth of knowledge here. I have a question for all you trading professionals/experts,

I've been developing a trading systems and after backtesting it on a universe of 300 hundred stocks (100 Nasdaq, 100 NYSE, & 100 AMEX), over a period of 1 year, I have a profit/loss ratio of 73%. But the average gain is 0.54% and the average loss is -0.75%. The overall average p/l is -0.16%. This is an interday swing that takes about 2 days.

Now cursively I think my system has a negative expectancy, but from what I've been reading here, with the right money management I can make it profitable. Does anyone have any thoughts on this? I'm relatively new with developing trading systems...

Thanks,
Jenna
 
It would help if you explained what your exit criteria and stop loss points, if any, are. Also, how exactly did you come up with your expectancy? I get a positive expectancy from your data but maybe I'm not understnading you accurately.
 
When you put in a tighter stop, reversing the average percent winner/loser ratio, does you win/loss ratio go dramatically down, making the system less profitable?

If so, it probably won't work. If you are not accounting for slippage and commissions, it definitely won't work.
 
may be you could give a more detailed description of your system... not how it works.. how you exit.

basically when you have a system and it gives you a negative performance on back-tested data... this should indicate to you that there is probably a better way. Maybe with a very good money management you can make it profitable... but if you would have a system with a positive outcome and put in this system all your energy for money management you can get much more out of it.

What is backtesting for: see if the system has been profitable in the past... and if it wasn't ask yourself why should it be provitable in the future?

alain
 
Actually, I didn't do that detailed of an test. Essentially, I ran the system over a time/stocks and came up with the days on which my system said to transact. I simulate a transact price which I'm sure I can get a fill on. I then use a default exit of a specific time period. 73% of the time I would be making a profit, which an average profit being 0.54%. The average loss for the remaining 27% is about -0.73%. This is assuming that there's no money management and no stop-loss orders. I merely wanted to see whether the system would get the direction of the market right, which it is 73% of the time. Now I'm looking to develope a money management technique for it - but first I'd like to hear your opinion as to whether a system like this can be helped with a good money management system or should I simply give it up because the average winning is less than the average loss.

Thanks again,
Jenna
 
what is your system based on...
try to adjust the parameters and then see if you can improve the profitable trades. if it is a trendfollowing system... try to adjust the indicators... if you for example see that it half of trades that are negative are generated in a different market environment try to add an indicator that excludes them. for example if you would use a macd.. combine that with an oscillator... very basic idea of optimizing a system.
if you have a break out system.. look where the falls break outs appear... is there a particular market situation and maybe you can combine the break out with an ADX for example so you can get rid of the negative trades.
Beside that for the beginning take a fixed stop strategy... a certain amount of money. and then by and by when you know your system set individuals stops... and make trailing stops. (this should be clear)

these are just some basic system trading approaches... on how to optimize a simple system. I don't know what kind of system or technique you are using so I can't give you detailed description.

alain
 
You must log in or register to reply here.
Back
Top