Reducing Factor-Portfolio Turnover Near a Selection Cutoff

Summary

The document describes a long–short equity factor strategy that ranks stocks by a combined daily factor score. At the next market open, it holds the most positive and most negative scores, while positions are fully liquidated as soon as a stock falls outside the selection boundary and fully re-established if it later qualifies again. This causes repeated trading among stocks near the cutoff.

The question seeks a way to reduce that turnover, but the provided text contains no answer or proposed method. It therefore identifies a portfolio construction problem rather than demonstrating a particular solution. Any practical response would need to weigh trading costs and position persistence against the intended score-based exposure; the document supplies no empirical tests, turnover estimates, or performance evidence for such choices.

Key ideas

• The strategy selects long and short stocks using the extremes of a combined factor score.
• Fully exiting and re-entering positions at a hard ranking boundary can create repeated trades.
• The question raises turnover near the cutoff but provides no solution or evidence.
• Any adjustment would involve trade-offs between turnover, costs, and score-based exposure.

Tags

# Excessive trading due to sharp cutoffs


# Excessive trading due to sharp cutoffs












I am running a stock trading system based on traditional factors (value, momentum, etc). I generate a combined factor score for each stock on every day at the close, and at the open of the next day, I trade to hold positions in the top 30% absolute value scores (long the most positive 15%, and short the most negative 15%).

Currently, if a stock drops out of the top 30% strongest absolute scores, I liquidate the position. Then if it re-enters, I have to trade back into the whole position. The result is that there's a lot of trading activity around the 70th percentile:

Is there a smarter way to do this?

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

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