Say you have a strategy that you found backtests well with 30 stocks. You then forward test the strategy on those 30 stocks and find that:
- 10 work well
- 10 don't work at all
- 10 break even
Don't go and trade the 10 profitable stocks, for this positive result is what would have happened with any pseudo-random data if trades follow some sort of normal distribution...
From experience:
================
If you have one strategy that works well in backtest across many similar stocks indicies (with the same parameters), and then performs similarly in walk forward across the majority of those similar stocks or indicies with >300 trades for position, >1000 trades for daytrades, this is significant enough...
- 10 work well
- 10 don't work at all
- 10 break even
Don't go and trade the 10 profitable stocks, for this positive result is what would have happened with any pseudo-random data if trades follow some sort of normal distribution...
From experience:
================
If you have one strategy that works well in backtest across many similar stocks indicies (with the same parameters), and then performs similarly in walk forward across the majority of those similar stocks or indicies with >300 trades for position, >1000 trades for daytrades, this is significant enough...
