Higher Moments: Skewness and Kurtosis
You'll learn how the third and fourth moments of a distribution reveal its asymmetry and tail behavior.
What Are Higher Moments?
You already know the first moment (expectation) tells you the center of a distribution, and the second central moment (variance) measures spread. The third and fourth standardized central moments—skewness and kurtosis—tell you about the shape of the distribution beyond just location and scale.
Skewness: Measuring Asymmetry
Skewness is the third standardized moment, defined as:
where μ is the mean and σ is the standard deviation.
- Skewness = 0: Symmetric distribution (like the normal distribution)
- Skewness > 0: Right-skewed (positive skew)—the right tail is longer, mass concentrates on the left
- Skewness < 0: Left-skewed (negative skew)—the left tail is longer, mass concentrates on the right
Think of it as asking: "Is there more extreme data on one side than the other?"
Kurtosis: Measuring Tail Heaviness
Kurtosis is the fourth standardized moment, defined as:
- Kurtosis = 3: Normal distribution (reference point)
- Kurtosis > 3: Heavy tails (leptokurtic)—more probability in extreme values
- Kurtosis < 3: Light tails (platykurtic)—less probability in extremes
Often we use excess kurtosis = Kurtosis − 3 so that zero corresponds to the normal distribution.
Kurtosis answers: "How prone is this distribution to extreme outliers?"
Key Takeaway: Skewness (third moment) measures asymmetry, while kurtosis (fourth moment) measures tail heaviness—together they describe distribution shape beyond mean and variance.