Sample Selection Bias

Sample selection bias arises when the data used to train a model is not representative of the broader population or the environment where the model will operate. In the context of financial derivatives, this often happens when a model is trained only on bull market data, failing to account for the structural changes that occur during a liquidity crisis.

If the training sample excludes specific market regimes or asset classes, the resulting model will lack the necessary intelligence to handle those conditions. This leads to poor performance when the market inevitably shifts or enters a period of high volatility.

Financial analysts must carefully curate their datasets to ensure they capture a wide spectrum of market behaviors. Addressing this bias is fundamental to building models that can survive the full cycle of financial history.