Skewness measures a distribution's asymmetry: positive when the right tail is longer, negative when the left one is. Kurtosis measures how heavy the tails are compared with a normal distribution, which has kurtosis 3, or excess kurtosis 0.
Card 1 of 244. Answer: Skewness measures a distribution's asymmetry: positive when the right tail is longer, negative when the left one is. Kurtosis measures how heavy the tails are compared with a normal distribution, which has kurtosis 3, or excess kurtosis 0.
About · Quizzes · History of AI · To suggest new questions or report an error send me a dm.
Skewness is the third standardized moment, E[(X − μ)³]/σ³, and kurtosis the fourth, E[(X − μ)⁴]/σ⁴. Kurtosis is often described as "peakedness", but it is driven almost entirely by the tails: high kurtosis means extreme values are more common than a normal distribution predicts.
Examples: salaries are right-skewed, so the mean sits above the median because a few very high salaries pull it up. The exponential distribution has skewness 2 and excess kurtosis 6; the uniform distribution has skewness 0 and excess kurtosis −1.2, lighter tails than a normal. Note that pandas' kurt() and SciPy's kurtosis() return excess kurtosis by default, so a normal sample scores near 0, not 3.
Why interviewers care: a skewed target often fits better after a log transform, and skew makes the mean a poor summary. Heavy tails mean outliers are part of the distribution, not errors, and models that assume normal tails, like many risk models, underestimate extreme events. Both statistics are noisy in small samples, especially kurtosis, which a single outlier can dominate.
