Volatility's Dirty Little Secret

Volatility is a mysterious creature in the world of finance. It's often touted as a key factor in determining investment success, but how well do we really understand it? A recent discussion on the Wilmott Forums sheds light on some surprising insights that every investor should know.

A thread titled "the 'thick' option trader - good exercise for quant student" delved into the intricacies of volatility estimation. The conversation centered around three main issues: nuisance parameters, Brownian motion, and risk-neutral vs. actual probability measures.

Estimating Volatility's Elephant in the Room

The first issue discussed was the problem of nuisance parameters. It turns out that estimating volatility requires knowing the expected return, which can be a challenge. If we use the actual return as an estimate, we might end up with unreasonable results. This is where Bayesians come in – they have a more elegant solution to this problem.

Bayesians don't rely on observed parameters as estimates of underlying ones. Instead, they compute a reasonable volatility estimate using Bayesian methods. This approach may seem abstract, but it highlights the importance of understanding the underlying assumptions in volatility estimation.

Brownian Motion's Reality Check

The second issue revolves around Brownian motion and its relation to volatility. In theory, volatility can be observed directly, making it impossible for a realized path of a Brownian motion to have a different volatility than the underlying one. However, when dealing with discrete points from the path, things get complicated.

The Volatility Paradox

The thread also touched on the difference between risk-neutral and actual probability measures. While this distinction might seem minor, it has significant implications for option pricing. Any realized volatility will differ under these two measures unless the expected return is the risk-free rate.

This paradox highlights a fundamental issue in volatility estimation: our understanding of Brownian motion is still evolving. It's not a mathematical contradiction, but rather a practical one that stems from our incomplete knowledge.

Portfolio Implications

So what does this mean for portfolios? For investors like those who follow the S&P 500 (C), Emerging Markets ETFs (EEM), or Goldman Sachs (GS), Bank of America (BAC), and Morgan Stanley (MS), understanding volatility is crucial. A better grasp of Brownian motion and risk-neutral vs. actual probability measures can help mitigate risks.

Actionable Takeaway

In conclusion, investors should be aware of the nuances in volatility estimation. By acknowledging the limitations of current methods and the need for further research, we can develop more accurate models. This, in turn, will allow us to make better investment decisions and optimize portfolio performance.