Hello Murray Ruggiero, I have enjoyed reading your articles over the past several years. Is it not true though that if you have a winning robust system that you are not going to share it with the world? Why then are there so many people touting systems and trade signal services?
how about this. a system that has never worked in the past will most likely never work in the future. why is that not a true statement and please explain.
It is true that systems which lost money in a backtest , could make money during some period in the future. I would not trust a system which failed a backtest because we do not have any evidence that it will work in the future. If you don't have this evidence it would not be a good idea to risk your money using that system.
There is NO guarantee of real time success regardless of backtesting performance. Just because some model (that was more than likely curb fitted or tested on a limited set of data points / or more than likely the "BEST" performing data points) does NOT mean that it actually works or that YOU have the ability to execute it.
I totally agree with you, if a system is curve fitted then a good backtest means nothing. The point I am making is that if you can't even get a good backtest you should not trust the system.
Yes you can use me as a reference. Enter Murray for the Referrers Name and mgf1205 for the Referral Code. If you purchase TradersStudio between 12/4/05 and 12/31/05 , you can fill out the rebate form on the site. http://tradersstudio.com/Default.aspx?tabid=131 * You need to be logged in to see the form You will receive a $100.00 rebate which will be credited to the credit card you used to place the order. This means for a limited time TradersStudio only cost you $399.00+shipping
How do you get an"" evidence that it will work in the future "",? From profitable back test? Is there EVER any evidence that system will work in the future?
1) First, if a system fails a backtest we can't trust it in the future 2) If it passes a system backtest we can do more analysis to measure the likelihood it will work in the future. This is all we can do with our backtest, the issue you are forgetting is design. Your system need to be based a valid theory , one which makes sense and you understand. Also if you understand why your system works , you also know if the markets changed to a point to break it. This concept is called premised based system development.