Understanding transaction cost models in competition scoring
Can someone explain how the transaction cost model works in the scoring engine? I see my gross Sharpe is 2.1 but net Sharpe drops to 1.4 after costs. That seems like a lot of slippage.
Are costs linear or do they scale with order size? Is there a market impact component?
14 Replies
This is a common pitfall. Make sure your features are computed before the prediction date, not on it. That's subtle look-ahead bias.
One thing to watch out for: survivorship bias in the training data. Make sure you include delisted securities.
Interesting thread! I've been exploring reinforcement learning for portfolio allocation. The challenge is defining the right reward function.
Market regimes are the elephant in the room. A strategy that works in a trending market will fail in mean-reverting conditions.
For those new to the platform: start with the tutorial competition. It has a smaller dataset and more forgiving scoring.
Be careful with the Kelly criterion for position sizing. Full Kelly is way too aggressive. I use quarter-Kelly in practice.
Have you considered using PCA to reduce the dimensionality of the feature space? I found that the first 10 components capture 80%+ of the variance.