I copied and tried optunity's 'grid search' option. What I found is that the searching term between parameter values are decided by num_evals. (High num_evals => 1..2..3..n/ Low num_evals => 1..5..10..n) That's why grid search is faster than the theory.
Posts made by woung717
RE: Genetic Optimization
Unrealistic backtesting return
First of all, thank you for open this amazing backtest platform for free.
I've have made several quant strategies using bt.
Sometimes I think backtesting result(return) is quite higher than what I expected.
Even when I implement a strategy wrong, it makes profits.
Usually, I rebalance 2 or 3 times annually and use momentum and volatility(1 / STDDEV of ROCP) to filtering or ranking.
I think there are errors in my data or it's just an equal-weighted portfolio effect.
Is there any way to audit the result except comparing it with the benchmark index?
Thank you :)