How Long do I Have to Backtest the Online Daytrading System?
Windowofworld.com – I am often asked how long it takes to backtest the online daytrading system. While there are no easy answers, I will give you some guidelines. There are several factors that you need to consider when determining the backtest period of your online daytrading system:
How many trades per day does your day trading system make? It doesn’t matter how long you test the day trading system; it is important that you receive enough trades to make statistically valid assumptions *: If your online day trading system makes three trades per day, i.e. 600 trades per year, then one year of testing provides you with enough data to make reliable assumptions *. But if your trading system only makes three trades per month, that is 36 trades per year, then you will have to test back several years to receive reliable data.
The underlying contract
You must consider the characteristics of the underlying contract. The chart below shows the average daily volume of the S&P e-mini:
It didn’t make sense to retest the trading system for the S&P e-mini before 1999, because the contract didn’t exist! In my opinion, it didn’t make sense to backtest the e-mini trading system before 2002 because at that time the markets were so different; lack of liquidity and different market participants. I believe that the reliable test period for the e-mini S&P is 2002 – 2004.