Moment Coefficient of Kurtosis Calculator

Use this unified calculator to find the moment coefficient of kurtosis for both ungrouped (raw) data and grouped (frequency distribution) data. Measure how peaked or flat a distribution is compared to the normal distribution.

Quick Start

Choose your data type, enter your values, and click Calculate:

Moment Coefficient of Kurtosis Calculator
Data Type Ungrouped (Raw Data) Grouped (Frequency Distribution)
Enter the X Values (Separated by comma,)
Type of Frequency Distribution DiscreteContinuous
Enter the Classes for X (Separated by comma,)
Enter the frequencies (f) (Separated by comma,)
Results
Number of Observations (N):
Mean of X values:
First Central Moment (μ₁):
Second Central Moment (μ₂):
Third Central Moment (μ₃):
Fourth Central Moment (μ₄):
Coefficient of Kurtosis (β₂):
Coefficient of Kurtosis (γ₂):

Understanding Kurtosis

Kurtosis (from Greek “kurtos” = curved) measures the degree of peakedness and tail heaviness of a distribution:

  • ✅ Compares shape to normal distribution
  • ✅ Measures tail behavior (propensity for outliers)
  • ✅ Captures distribution peakedness
  • ✅ Two forms: β₂ (algebraic) and γ₂ (excess, practical)

Three Distribution Types

Type β₂ γ₂ Shape Tails
Leptokurtic > 3 > 0 Peaked/Sharp Heavy
Mesokurtic = 3 = 0 Normal Moderate
Platykurtic < 3 < 0 Flat Light

Formulas

Coefficient of Kurtosis (β₂ Form)

$$\beta_2 = \frac{\mu_4}{\mu_2^2}$$

Where:

  • $\mu_4$ = Fourth central moment
  • $\mu_2$ = Second central moment (variance)

Coefficient of Kurtosis (γ₂ Form) - Excess Kurtosis

$$\gamma_2 = \beta_2 - 3$$

Or directly:

$$\gamma_2 = \frac{\mu_4}{\mu_2^2} - 3$$

Why Subtract 3?

The normal distribution has β₂ = 3, so:

  • γ₂ = 0: Distribution has normal kurtosis (mesokurtic)
  • γ₂ > 0: More peaked than normal (leptokurtic)
  • γ₂ < 0: Flatter than normal (platykurtic)

How to Use This Calculator

For Ungrouped (Raw) Data

Step 1: Select “Ungrouped (Raw Data)”

Step 2: Enter your data values separated by commas

Step 3: Click “Calculate”

Results will show:

  • Mean and all central moments
  • Both kurtosis coefficients (β₂ and γ₂)
  • Distribution classification

For Grouped (Frequency Distribution) Data

Step 1: Select “Grouped (Frequency Distribution)”

Step 2: Choose distribution type and enter data

Step 3: Click “Calculate”


Distribution Types Explained

Leptokurtic (Peaked)

  • β₂ > 3 or γ₂ > 0
  • Sharp peak, pronounced central concentration
  • Heavy tails (more outliers)
  • Examples: Financial returns, stock prices (frequent extremes)

Mesokurtic (Normal)

  • β₂ = 3 or γ₂ = 0
  • Moderate peak, typical bell curve
  • Normal tail behavior
  • Examples: Many natural phenomena, test scores

Platykurtic (Flat)

  • β₂ < 3 or γ₂ < 0
  • Flat distribution, spread out
  • Light tails (fewer outliers)
  • Examples: Uniform distributions, rainfall distribution

Worked Examples

Example 1: Ungrouped Data - Normal Distribution

Data: Values: 3, 5, 7, 9, 11, 13, 15

Find the coefficient of kurtosis.

Solution:

Step 1: Calculate Mean

$$\overline{x} = \frac{3+5+7+9+11+13+15}{7} = 9$$

Step 2: Calculate Central Moments

Deviations: -6, -4, -2, 0, 2, 4, 6

μ₁ = 0 μ₂ = (36+16+4+0+4+16+36)/7 = 112/7 = 16 μ₃ = (-216-64-8+0+8+64+216)/7 = 0/7 = 0 μ₄ = (1296+256+16+0+16+256+1296)/7 = 3136/7 = 448

Step 3: Calculate Kurtosis

$$\beta_2 = \frac{448}{16^2} = \frac{448}{256} = 1.75$$

$$\gamma_2 = 1.75 - 3 = -1.25$$

Answer: β₂ = 1.75, γ₂ = -1.25

Interpretation: The distribution is platykurtic (flatter than normal) with γ₂ < 0. This uniform distribution has a flat shape and light tails.


Example 2: Ungrouped Data - Peaked Distribution

Data: Values: 5, 5, 5, 5, 5, 10, 20

Find the coefficient of kurtosis.

Solution:

Step 1: Calculate Mean

$$\overline{x} = \frac{5+5+5+5+5+10+20}{7} = 8.57$$

Step 2: Calculate Central Moments

Deviations: -3.57, -3.57, -3.57, -3.57, -3.57, 1.43, 11.43

μ₂ = (12.74+12.74+12.74+12.74+12.74+2.04+130.64)/7 = 43.29 μ₃ = (-45.53-45.53-45.53-45.53-45.53+2.92+1493.08)/7 = 122.34 μ₄ = (162.5+162.5+162.5+162.5+162.5+4.18+17045.5)/7 = 2661.16

Step 3: Calculate Kurtosis

$$\beta_2 = \frac{2661.16}{43.29^2} = \frac{2661.16}{1874}{} = 1.42$$

Actually let me recalculate: β₂ = 2661.16/(43.29)² = 2661.16/1874.22 = 1.42

$$\gamma_2 = 1.42 - 3 = -1.58$$

Answer: β₂ ≈ 1.42, γ₂ ≈ -1.58

Interpretation: Despite concentration, this distribution is still platykurtic due to extreme values causing high variance.


Example 3: Grouped Data (Discrete) - Test Scores

Problem: Score distribution for 50 students. Calculate kurtosis.

Score 3 4 5 6 7
Frequency 5 10 20 10 5

Solution:

Step 1: Calculate Mean

$$\overline{x} = \frac{3(5)+4(10)+5(20)+6(10)+7(5)}{50} = \frac{250}{50} = 5$$

Step 2: Calculate Central Moments

Deviations: -2, -1, 0, 1, 2

μ₂ = [5(4)+10(1)+20(0)+10(1)+5(4)]/50 = 60/50 = 1.2 μ₃ = [5(-8)+10(-1)+20(0)+10(1)+5(8)]/50 = 0/50 = 0 μ₄ = [5(16)+10(1)+20(0)+10(1)+5(16)]/50 = 180/50 = 3.6

Step 3: Calculate Kurtosis

$$\beta_2 = \frac{3.6}{1.2^2} = \frac{3.6}{1.44} = 2.5$$

$$\gamma_2 = 2.5 - 3 = -0.5$$

Answer: β₂ = 2.5, γ₂ = -0.5

Interpretation: The distribution is platykurtic (flatter than normal). Scores are distributed symmetrically with a flatter shape than a normal distribution.


Example 4: Grouped Data (Continuous) - Age Distribution

Problem: Age distribution of 60 people. Calculate kurtosis.

Age Group 20-30 30-40 40-50 50-60 60-70
Frequency 3 8 35 10 4

Solution:

Step 1: Use Midpoints and Calculate Mean

Midpoint Frequency f×x
25 3 75
35 8 280
45 35 1,575
55 10 550
65 4 260
Total 60 2,740

$$\overline{x} = 2740/60 = 45.67$$

Step 2: Calculate Central Moments

Deviations: -20.67, -10.67, -0.67, 9.33, 19.33

μ₂ = [3(427.7)+8(113.9)+35(0.45)+10(87.0)+4(373.6)]/60 = 3584/60 = 59.73

μ₄ = [3(183.5k)+8(13k)+35(0.2)+10(7.6k)+4(139k)]/60 = 1,052,000/60 = 17,533

Step 3: Calculate Kurtosis

$$\beta_2 = \frac{17533}{59.73^2} = \frac{17533}{3567.7} = 4.91$$

$$\gamma_2 = 4.91 - 3 = 1.91$$

Answer: β₂ ≈ 4.91, γ₂ ≈ 1.91

Interpretation: Highly leptokurtic distribution (γ₂ > 1). The sharp concentration of ages around 40-50 creates a peaked distribution with heavy tails relative to normal.


Relationship Between Skewness and Kurtosis

Skewness Kurtosis Distribution Shape
0 0 Normal (Bell curve)
0 > 0 Peaked, symmetric
0 < 0 Flat, symmetric
≠ 0 0 Skewed, normal tails
≠ 0 ≠ 0 Both skewed and unusual peak

Kurtosis Applications

Financial Markets

  • High kurtosis: Stock returns show frequent extreme movements
  • Management: Risk must account for tail behavior
  • Trading: Adjust strategies for outlier probability

Quality Control

  • High kurtosis: Process produces occasional defects
  • Low kurtosis: Process very stable and consistent

Medical Research

  • Normal kurtosis: Drug response follows expected pattern
  • High kurtosis: Some patients have extreme reactions

Common Mistakes to Avoid

WRONG: Confusing β₂ and γ₂ ✓ RIGHT: β₂ uses normal as 3; γ₂ uses normal as 0 (excess kurtosis)

WRONG: Thinking high kurtosis always means “bad” ✓ RIGHT: Context matters;financial risk vs. quality control trade-offs

WRONG: Ignoring skewness when interpreting kurtosis ✓ RIGHT: Consider both skewness and kurtosis together

WRONG: Using only outliers to judge kurtosis ✓ RIGHT: Kurtosis is about all deviations, not just extremes


Key Differences: Ungrouped vs. Grouped Data

Aspect Ungrouped Grouped
Precision Exact moments Approximate from midpoints
Calculation Direct from values Using class midpoints
Data Loss None Some from grouping
Computational More values Simplified with frequencies

Visual Interpretation

Leptokurtic (Peaked, Heavy Tails)

  • Sharp central peak
  • Tails don’t drop off as quickly
  • More mass in extremes
  • Better for extreme value analysis

Mesokurtic (Normal, Moderate Tails)

  • Bell curve shape
  • Standard tail behavior
  • Reference standard
  • Common in nature

Platykurtic (Flat, Light Tails)

  • Spread out, flat top
  • Tails drop off quickly
  • Few extreme values
  • Uniform-like behavior

Related shape measures:

Related statistics:

Learn More: