The median deviation of the predicted expected-1-lot lots from its realized value, within Pool B (applications above HKD 5m).
Methodology
Expected 1-lot lots = the smallest tier whose expected lots ≥ 1; relative error = |predicted − actual| ÷ actual, then take the median. Predictions are made at each stock's realized oversubscription.
Current Value
0.0%
Sample Size
163
Methodology
This metric comes from a walk-forward backtest on stocks listed in 2025 and later — the anchors (market regime, index, sponsor heat, first-day P&L) use only data visible before each stock listed; the regression coefficients are fitted on the full history, matching production prediction.
This page documents the definition, methodology and current value of the prediction metric "Pool B Expected-1-Lot Median Relative Error". The value is aggregated from the prediction_history table and refreshed by the backfill script, for objectively assessing model accuracy — not investment advice.