Evidence of Predictable Behavior of Security Returns
Read the paperopens doi.org in a new tab
What they found
Jegadeesh documented the one-month reversal effect: for U.S. stocks from 1934 to 1987, a stock's return last month negatively predicts its return this month, while returns twelve months ago positively predict it. A strategy of buying last month's losers and selling last month's winners earned a striking abnormal return of around 2% per month before costs. This short-term reversal is the reason momentum researchers skip the most recent month when building their signal.
What you can use
- The most recent month's return tends to reverse; the prior year's return tends to continue.
- This is why standard momentum uses the '12 minus 1' month formation window.
- Large paper profits from one-month reversal are mostly consumed by bid-ask bounce and trading costs in practice.
Caveats
Very high turnover strategy; later work shows most of the profit is in small, illiquid stocks and shrinks dramatically after realistic costs. Sample ends 1987.
Tags: mean-reversion, short-term-reversal, equities
Summaries are our own reading of the paper, not the authors' words. Educational only, not advice. Discuss it in Book Club.