Momentum Variations

Does Complexity Beat Simplicity?

August 2018. Reading Time: 10 Minutes. Author: Nicolas Rabener.

SUMMARY

  • The simplicity of the Momentum factor can be intellectually challenging
  • Various alternative Momentum versions highlight remarkable similar return profiles
  • The robustness is an attractive characteristic of the investment strategy

INTRODUCTION

What do selfies, the Kardashians, Crocs, blue cheese, and Boris Johnson have in common? They all rank within the top 50 things that split the opinion of modern British people according to a recent survey. An equivalent list in finance likely includes cryptocurrencies, the valuation of the S&P 500, the direction of interest rates, the future of the Euro, and the Momentum factor. The latter is a slightly more unusual component as there is a significant amount of academic research that has shown that Momentum, both time series and cross-sectional, works within various asset classes, across different markets and time periods.

However, especially long-short Momentum in equities tends to be either favoured, mostly by quants, or actively disliked, usually by discretionary investors. It is reasonable that investors who create detailed fundamental models, regularly meet corporate CEOs, and often studied hundreds of hours after work for the charted financial analyst (CFA) designation struggle with an investment strategy that can be replicated by high school students. In this short research note, we will therefore investigate more complex variations of the Momentum factor, which might intellectually be more satisfying and potentially result in higher risk-adjusted returns (read Momentum Factor: Intra vs Cross-Sector).

METHODOLOGY

We focus on the Momentum factor in the US, Europe and Japan, which is defined as buying the winning and shorting the losing stocks. The factor performance is calculated by constructing a long-short beta-neutral portfolio of the top and bottom 10% of stocks ranked by the factor definition. Only stocks with a minimum market capitalisation of $1 billion are included. Portfolios are rebalanced monthly and each transaction incurs costs of 10 basis points. We investigate the following Momentum variations:

  • Classic (Fama-French): Sorting stocks by their performance over the last 12 months, excluding the most recent month
  • Sh