Short-term reversal

Short-term reversal is the cross-sectional pattern in which stocks that fell hardest over a recent month tend to bounce, and recent winners tend to give some of it back. It is the opposite horizon from the multi-month momentum effect. Jegadeesh documented predictable monthly reversals in individual stocks. The usual portfolio is long recent losers and short recent winners, then reset.

Not investment advice. Nothing here is a recommendation to buy or sell anything. Backtests and historical studies are not guarantees, and a rule that worked in one sample can fail in the next. Trading can lose money. Links are for education, not endorsements. Do your own research.

Rules

Each month, rank a liquid stock universe on the prior month's return, skipping the most recent few days if the implementation wants to reduce bid-ask bounce. Buy the worst decile and short the best decile in equal weight, or dollar-neutral weights. Hold about one month and re-rank. Exclude names that fail a price or volume floor so the short book is borrowable.

Where AI fits

AI can build the monthly ranking, apply borrow and cost assumptions, and keep a buffer so tiny rank changes do not force a full turnover. It can also check that the result is not just the bid-ask bounce Jegadeesh discussed. A model should not be allowed to pick the lookback after seeing the full-sample table.

Markets

Liquid individual stocks. Needs a universe large enough for deciles and a way to borrow shorts.

Timeframes

Monthly rebalance. The signal is a one-month return, not an intraday oscillator.

Risks

Recent losers are often distressed. Shorting recent winners fights the momentum effect if the lookback is lengthened by mistake. Transaction costs can erase a one-month rebalance. The pattern is not the same thing as an RSI dip on a single chart.

← Back to all strategies