mesokurtosis

Mesokurtosis is a term used in statistics to describe the similarity between a distribution’s kurtosis and that of the normal distribution. Kurtosis measures the heaviness of tails and the sharpness of a distribution’s peak relative to a normal curve. A mesokurtic distribution has a kurtosis close to that of the normal distribution, with an excess kurtosis near zero. In practice, this means the distribution’s tails are neither particularly heavy nor unusually light, and its peak is not excessively pronounced.

Kurtosis classifications contrast mesokurtic with leptokurtic, which have sharp, tall peaks and/or heavy tails, and platykurtic, which are flatter with lighter tails. The normal distribution is the standard reference and is typically described as mesokurtic. In terms of definitions, some texts use the Fisher definition (excess kurtosis), where mesokurtic corresponds to excess kurtosis around zero; under the Pearson definition, a kurtosis of 3 corresponds to the mesokurtic case.

Estimating mesokurtosis involves computing the sample kurtosis from data. Larger samples reduce sampling error, but the estimate can be strongly influenced by outliers and deviations from normality. In applied analysis, mesokurtosis is used to assess tail behavior and to guide the selection of statistical methods that assume near-normal tails.

Limitations include the fact that kurtosis is only one aspect of a distribution’s shape; different distributions can share similar kurtosis values. Therefore, mesokurtosis should be interpreted alongside skewness and other descriptive measures.