Cross-Sectional Multi-Factor Ranking for Long-Short Crypto Trading

Summary

This announcement outlines a multi-factor strategy for OKX that ranks a universe of coins by a composite factor value and divides them into three roughly equal groups. It proposes going long the lowest-ranked group and short the highest-ranked group. If larger factor values are associated with higher returns, the ranking direction can be reversed or the factor transformed. The method therefore describes a cross-sectional long-short portfolio based on factor sorts.

Factor quality is assessed roughly through backtested return and Sharpe ratio, alongside checks for expression simplicity and sensitivity to group size, rebalance spacing, and the backtest start date. In live operation, the strategy is said to rebalance in real time or at short intervals, closing positions when group membership calls for taking the opposite side. The announcement also states that the strategy cannot be backtested through its interface, creating a tension with its proposed evaluation process. It gives no factor definition or performance results and warns that one-sided markets and large positions can lead to liquidation.

Key ideas

• The strategy sorts coins by a composite factor and forms three approximately equal groups.
• It proposes longing the lowest factor group and shorting the highest, with direction reversed when the factor-return relationship calls for it.
• Factor evaluation considers return, Sharpe ratio, expression simplicity, and sensitivity to portfolio and testing choices.
• Live rebalancing is described as frequent, with exits when group changes require a position in the opposite direction.
• The announcement supplies no factor formula or results and says the strategy itself cannot be backtested.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.