Backtest
Running a strategy's rules on historical data and measuring the results. Gives an idea of expectancy and drawdown before going live.
A backtest is done by hand (stepping bar by bar) or with code. Spread, commission and slippage must be included; otherwise the result is optimistic. You need a sufficient sample (at least 100-200 trades, across different market regimes).
A backtest is no guarantee of the future; against the risk of overfitting, out-of-sample testing and a forward test are a must.
Related terms
Where it appears in the guides
Guides whose text uses this term; most frequent first.
- İleri test (forward test) ve demo dönemiStrategy & backtesting
- Algoritmik trading'e giriş: Python ile ilk adımlarTrading systems & data
- Backtest: veri, örneklem büyüklüğü ve hatalarStrategy & backtesting
- Aşırı uyum (overfitting) ve sağlamlık testleriStrategy & backtesting
- Kural tabanlı strateji nasıl yazılırStrategy & backtesting
- Piyasa verisi kaynakları: tick, dakika ve günlükTrading systems & data