Skip to content
All library documents

Constructing Fama–French Short- and Long-Term Reversal Factors

Article Quant Q&A · Author: Mostafa Bouzari
Read the original

Summary

The document describes how to construct the Fama–French short-term and long-term reversal factors from six value-weighted portfolios. For each factor, stocks are sorted monthly into two size groups using the NYSE median market equity breakpoint and three groups by prior return using the 30th and 70th NYSE percentiles. The short-term sort uses prior one-month returns, while the long-term sort uses returns from months 13 through 60.

Each reversal factor is the average return of the low-prior-return portfolios minus the average return of the high-prior-return portfolios, averaging across the small and big groups. This provides a portfolio-level recipe for forming the factors, which the questioner wants to use in return regressions alongside other equity factors. The explanation is a concise construction specification rather than a full replication guide: it does not discuss data cleaning, missing observations, portfolio rebalancing details beyond monthly formation, or how to test the resulting factor series.

Key ideas

  • Both reversal factors use six value-weight portfolios formed at monthly intervals.
  • The sorts combine two size groups with three groups ranked by prior returns.
  • Short-term reversal uses the prior one-month return sort, while long-term reversal uses months 13 through 60.
  • Each factor subtracts the average return of high prior-return portfolios from that of low prior-return portfolios.

Tags

EquitiesFactor investingMean reversionPortfolio constructionUS markets
Full text
# How to calculate FF short/long term reversal for a portfolio?


# How to calculate FF short/long term reversal for a portfolio?












As a newcomer to the finance world, I am working on replicating the Fama-French 5-factor model, along with Momentum and Short/Long-Term Reversal factors, to regress my portfolio's industry-adjusted returns against these factors. My goal is to determine whether these factors can explain my returns or if I can achieve significant alphas.

I have already successfully replicated the Fama-French 3-factor and 5-factor models, as described in their original papers (Fama & French, 1993; Fama & French, 2015). Similarly, I have implemented the Momentum factor (Carhart, 1997) and the HML factor based on current market equity, following the insights from the "Devil in HML Details" paper (Asness & Frazzini, 2013).

However, I have been unable to locate the original paper that introduces the Short/Long-Term Reversal effects, and consequently, I haven't been able to replicate this factor.

If possible, I would appreciate it if someone could point me to the original paper introducing the Short/Long-Term Reversal effect. Additionally, I would be grateful for guidance on how to calculate the monthly Short/Long-Term Reversal effect for my portfolio data.

## Answer by Mostafa Bouzari (score 2, accepted)

https://quant.stackexchange.com/a/80729

The answer lies in fama french's website for short-term and long-term reversal effects.

short-term:

> We use six value-weight portfolios formed on size and prior (1-1) returns to construct ST_Rev. The portfolios, which are formed monthly, are the intersections of 2 portfolios formed on size (market equity, ME) and 3 portfolios formed on prior (1-1) return. The monthly size breakpoint is the median NYSE market equity. The monthly prior (1-1) return breakpoints are the 30th and 70th NYSE percentiles. ST_Rev is the average return on the two low prior return portfolios minus the average return on the two high prior return portfolios, ST_Rev = 1/2 (Small Low + Big Low)- 1/2(Small High + Big High).

Long-Term:

> We use six value-weight portfolios formed on size and prior (13-60) returns to construct LT_Rev. The portfolios, which are formed monthly, are the intersections of 2 portfolios formed on size (market equity, ME) and 3 portfolios formed on prior (13-60) return. The monthly size breakpoint is the median NYSE market equity. The monthly prior (13-60) return breakpoints are the 30th and 70th NYSE percentiles. LT_Rev is the average return on the two low prior return portfolios minus the average return on the two high prior return portfolios, LT_Rev = 1/2 (Small Low + Big Low) - 1/2(Small High + Big High).

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.