Tuesday, November 20, 2012

Another Blow to Modern Portfolio Theory

From Systematic Relative Strength:
Modern Portfolio Theory is predicated on the ability to construct an efficient frontier based on returns, correlations, and volatility.  Each of these parameters needs to be accurate for the efficient frontier to be accurate.  Since forecasting is tough, often historical averages are used.  Since the next five or ten years is never exactly like the last 50 years, that method has significant problems.  Apologists for modern portfolio theory claim that better efficient frontiers can be generated by estimating the inputs.  Let’s imagine, for a moment, that this can actually be done with some accuracy.

There’s still a big problem.  Volatility bumps up during adverse market conditions, as reported by Research Affiliates.  And correlations change during declines—and not in a good way.
From the abstract of a recent paper, Quantifying the Behavior of Stock Correlations Under Market Stress:
Understanding correlations in complex systems is crucial in the face of turbulence, such as the ongoing financial crisis. However, in complex systems, such as financial systems, correlations are not constant but instead vary in time. Here we address the question of quantifying state-dependent correlations in stock markets. Reliable estimates of correlations are absolutely necessary to protect a portfolio. We analyze 72 years of daily closing prices of the 30 stocks forming the Dow Jones Industrial Average (DJIA). We find the striking result that the average correlation among these stocks scales linearly with market stress reflected by normalized DJIA index returns on various time scales. Consequently, the diversification effect which should protect a portfolio melts away in times of market losses, just when it would most urgently be needed.
I bolded the part that is most inconvenient for modern portfolio theory.  By the way, this isn’t really cutting edge.  The rising correlation problem isn’t new, but I find it interesting that academic papers are still being written on it in 2012....MORE
And that's a wrap.
 multi echelon inventory optimization