What does the term "kurtosis" primarily refer to in statistics?

Kurtosis is a statistical measure that describes the shape of a distribution's tails in relation to its peak. Specifically, it refers to the height and sharpness of the peak of the distribution. A distribution with high kurtosis has heavy tails and a sharper peak, indicating more data in the tails and a greater likelihood of extreme outcomes. Conversely, a distribution with low kurtosis tends to have flatter peaks and lighter tails, suggesting that values are more evenly spread out and extreme values are less likely.

Understanding kurtosis helps in risk assessment, as distributions with high kurtosis can signal the potential for outliers or extreme changes in a dataset, which is particularly important in fields like finance and risk management. The other options relate to different aspects of statistical distributions—symmetry, averages, and dispersion—each of which serves its own purpose in describing a dataset but does not specifically capture the essence of kurtosis.