Market impact models for realistic backtesting
Ignoring market impact in backtests leads to significantly inflated performance, especially for strategies that trade less liquid instruments.
Models to consider:
- Linear impact: cost proportional to trade size (simplistic)
- Square-root impact: cost ~ sqrt(volume/ADV), more realistic
- Almgren-Chriss: optimal execution framework
18 Replies
The documentation for the API is at /docs -- it's OpenAPI/Swagger format. Very helpful for understanding submission formats.
I disagree about the GARCH approach. In my experience, realized volatility estimators (like the Rogers-Satchell estimator) outperform parametric models.
Has anyone tried using attention mechanisms for this? The temporal attention weights could tell you which historical periods are most relevant.
The data quality in this competition is actually quite good compared to real-world datasets. In practice, you'd spend 60%+ of your time cleaning data.
Be careful with the Kelly criterion for position sizing. Full Kelly is way too aggressive. I use quarter-Kelly in practice.
Has anyone tried using attention mechanisms for this? The temporal attention weights could tell you which historical periods are most relevant.
The documentation for the API is at /docs -- it's OpenAPI/Swagger format. Very helpful for understanding submission formats.
I've found that sector neutrality is a key factor in the scoring. Strategies that are long one sector and short another tend to underperform.