Weighting Training Samples by Tradability
Giving a machine learning model's training loss more weight on observations that were actually easy to trade at the time, so it doesn't learn patterns that only existed in illiquid, hard-to-execute names.
A cross-sectional model trained to predict next-month stock returns treats every observation in its training set equally by default, whether that observation is a heavily traded large-cap or a thinly traded microcap that a real portfolio could barely have bought or sold without moving the price. If small, illiquid names disproportionately drive the largest apparent return anomalies — which they often do, because a handful of shares changing hands can swing a stock price sharply — an unweighted model can end up fitting patterns that a real strategy could never actually capture.
Weighting training samples by tradability means assigning each observation a weight (used in the loss function during training) that reflects how easily the position could have been traded at that point in time — typically some function of average daily dollar volume, bid-ask spread, or market capitalization, so illiquid observations count for less and liquid ones count for more. A stock trading $50 million a day might get a weight ten times that of one trading $500,000 a day, roughly mirroring the difference in position size a real fund could actually put on.
This is a training-time fix, distinct from simply excluding illiquid names outright: down-weighting still lets the model learn from the full cross-section (illiquid names carry real information about the broader pattern) while nudging the fitted model toward the signal that survives in the liquid universe a live strategy would actually trade.
Weighting training observations by tradability (dollar volume, spread, or market cap) keeps a cross-sectional model from over-learning from illiquid names that drive outsized apparent returns but can't actually be traded at that size, without discarding those observations entirely.
Further reading
- Practitioner notes on sample construction for cross-sectional equity models