Newsletter – September / October 2014 – Extreme reversion is not an outlier


Extreme reversion is not an outlier

Managing money with a strong or primary filter in the analysis of security price action has changed over the last 5 or 6 years. While the tenants or tools of trend identification remain fixed, proven over many decades of empirical evaluation, and while new indicators are continually being created, markets and their constituent securities dwell more and more at the extremes. The extreme nature of markets is also impacting a pure value or fundamental approach

Why does it matter? It matters because, liner regression analysis, a central resource for all managers, whatever their style and one of a few tools that broadly overlap technical and fundamental analysis, is less effective, diminished. Prior to high frequency or algorithmic trading, ubiquitous hedge fund market participation and declining trading volumes, standard linear regression helped define for managers – value, growth or momentum – whether a security was presently investible or not.

Why would a manager, whether they have a dominant technical or fundamental inclination, buy a security for client portfolios that is “overbought” or “overvalued”? In reality securities do not have the bias that Managers do, and they can exist at whatever price level they find equilibrium between buyers and sellers.

There is no necessary, eventual, justice in market action, no necessary reversion to the mean. They care not a whit whether anything makes sense.

Read Full Article (PDF)