(August 9, 2013) — If there have been securities traded in the US since 1792, why does most of the fund industry backtest to just 1927?
Christopher Geczy, from The Wharton School of the University of Pennsylvania, and Mikhail Samonov, founder of quantitative investment research firm Octoquant, decided to look back as far as possible in a recent paper—and made some discoveries about momentum strategies.
They first considered the backdrop to markets before 1927.
“The 19th and early 20th centuries are filled with expansions, recessions, wars, panics, manias, and crashes, all providing a rich out-of-sample history. Limiting studies to the post-1925 period introduces a strong selection bias and does not capture the full distribution of possible outcomes,” the authors began.
They took the case of price momentum as their first example and found that in the post-1925 period there had only been one case of a decade-long negative compound return. This occurrence was therefore considered to be an outlier and the rest of the period was considered normal.
That was the first mistake, the authors said. Markets entered another such period after the most recent crash, which affected portfolios using this strategy and got the authors thinking.
“The repeated underperformance raised practical questions about the outlier conclusion and what the actual distribution of momentum profits is. By extending the momentum data back to 1801, we create a more complete picture of the potential outcomes of momentum profits, discovering seven additional negative decade-long periods prior to 1925.”
As for the success of using the strategy to construct portfolios, the pair found the momentum effect was significant before 1925, but only it had only about half of the influence during the first period as in the latter one.
From 1801 to 1926, the equally weighted top third of stocks sorted on price momentum outperformed the bottom third by 0.28% per month, compared to 0.58% per month for the 1927-2012 period. Linking the two periods together generated a 212-year history of momentum returns, averaging 0.4% per month.
They found the strategy would have generated significant outperformance in both periods, but it is not a product of data-mining, rather it is highly variable over time.
They also discovered that the strategy was not just related to analysing single stocks, but considering a combination of common risk and specific company securities.
To read the full paper, including detailed formulae, click here.