Understanding Platykurtic Distributions: A Safe Investment Strategy
Platykurtic describes a statistical distribution with negative excess kurtosis. Essentially, distributions of this nature have thinner tails compared to normal distributions, leading to fewer extreme outcomes—whether exceptionally positive or negative. Contrast this with a leptokurtic distribution, which has positive excess kurtosis, indicating fatter tails and a higher propensity for extreme results.
Investors carefully examine these statistical characteristics when deciding where to allocate their resources. Those who are risk-averse typically lean towards platykurtic distributions, knowing they’re less likely to encounter extreme market swings.
Key Takeaways
- Platykurtic distributions feature negative excess kurtosis.
- These distributions have a lower likelihood of extreme events compared to normal distributions.
- Risk-averse investors could benefit from targeting investments with platykurtic returns to minimize large negative occurrences.
Deep Dive into Platykurtic Distributions
Statistical distributions can be categorized into three main types: leptokurtic, mesokurtic, and platykurtic. These categories vary by the amount of excess kurtosis each possesses, which directly influences the probability of encountering extreme events. For example, while mesokurtic distributions, including the well-known normal distribution, have a kurtosis of three, those with more than three exhibit ‘positive excess kurtosis’ and those with less, ‘negative excess kurtosis.’
Leptokurtic distributions show a higher possibility of extreme outcomes due to their positive excess kurtosis, but platykurtic distributions stand at the other end, characterized by a lower chance of such outcomes due to their negative excess kurtosis.
Take a look at these distributions depicted in two different figures. On the left, the immediate differences in tail behavior might be subtle, but to gain clear insights, the figure on the right uses a technique known as a quantile-quantile (Q-Q) plot. This method distinctly contrasts the differences.
Navigating Market Varieties
Most investors agree that equity market returns usually follow a leptokurtic distribution, which means that while average returns might be steady, there can be sudden significant deviations—commonly referred to as ‘black swan events’—that are markedly less predictable. Therefore, more cautious investors often avoid these markets to circumvent these unexpected extremes, instead opting for the safer boundaries of platykurtic returns. Conversely, some daring investors might seek leptokurtic markets hoping the occasional significant gains will surpass the extremes’ downswings.
Real-World Example of Platykurtic Distribution
Morningstar conducted a study examining the excess kurtosis levels for various assets from February 1994 to June 2011, including U.S. and international equities, real estate, commodities, cash, and bonds.
At the lower end of the excess kurtosis spectrum, cash and international bonds were ranked with figures of -1.43 and 0.58 respectively. Meanwhile, U.S. high-yield bonds and hedge-fund arbitrage strategies were on the opposite end, showing 9.33 and 22.59 excess kurtosis levels respectively.
Assets with middle-ground excess kurtosis included international real estate at 2.61, equities from emerging international economies at 1.98, and commodities at 2.29.
An investor evaluating this spectrum can quickly grasp which assets align with their risk preferences. Risk-averse individuals would likely gravitate towards low-kurtosis investments to avoid significant market swings, whereas more risk-tolerant investors might venture into high-kurtosis assets, embracing the potential for drastic returns.
Related Terms: leptokurtic distribution, mesokurtic distribution, kurtosis, black swan events, risk aversion.
References
- Morningstar. “The Real World Is Not Normal Introducing the New Frontier: an Alternative to the Mean-Variance Optimizer”.
- Morningstar. “The Real World Is Not Normal Introducing the New Frontier: an Alternative to the Mean-Variance Optimizer”, Page 2.
