Skip to content

Kurtosis Calculator

Enter a list of numbers and choose the sample or population method. The calculator finds the fourth standardised moment and subtracts 3 to report excess kurtosis. For the default eight values, sample excess kurtosis is 0.9406 and raw kurtosis is 3.9406, described as leptokurtic (heavy tails).

Updated 2026-11-19 Editorial estimate Calccc Editorial Team Free · No sign-up

Separate numbers with commas, spaces, or semicolons. At least four values are needed for the sample estimate.

Scenarios A/BSave two input sets and compare the results. Stored only in this browser.
AEmpty
BEmpty

Your inputs stay in this browser. No account, no uploads.

Professional worksheet

Detailed schedule & assumptions

Inputs used, method, sensitivity, and any year-by-year table. The summary above stays compact; the audit lives here.

Professional worksheet

Inputs, scope, and sensitivity

Inputs used

Method and scope

Result audit

MetricValue

What this calculator does

Kurtosis measures how much of a distribution sits in the tails rather than near the centre. A normal distribution has a kurtosis of 3, so statisticians usually subtract 3 and report excess kurtosis, which is zero for a normal distribution. Positive excess means heavier tails and a sharper peak; negative excess means lighter tails and a flatter top. This calculator accepts a list of values and computes both the population and bias-corrected sample versions, then reports the one you select. It also shows the mean, standard deviation, and variance so the result can be checked. Kurtosis is sensitive to extreme values, which is exactly the point: a single outlier can dominate the statistic and reveal a tail that the mean and standard deviation hide.

How to use it

  1. Enter the data values separated by commas, spaces, or semicolons.
  2. Use at least four values for the sample estimate; smaller samples fall back to the population formula.
  3. Choose the sample method when the data are a sample from a larger population.
  4. Choose the population method when the data are the whole population.
  5. Read excess kurtosis first; raw kurtosis is excess plus 3.
  6. Check the standard deviation and the outlier before interpreting the shape.

How it works

Raw kurtosis = E[(x − mean)⁴] ÷ variance². Excess kurtosis = raw kurtosis − 3. The sample version applies a bias correction based on n. For the default values, the variance is 4.00, the standard deviation is 2.00, and the sample excess kurtosis is 0.9406.

Example calculations

Example 1: Eight values with moderate tails

The mean is 5.00 and the standard deviation is 2.00. The sample excess kurtosis is 0.9406, so the raw kurtosis is 3.9406. That is above the normal value of 3, which is why the shape is described as leptokurtic, but with only 8 values the estimate is uncertain.

Example 2: A uniform-looking sequence from 1 to 10

The values 1 to 10 have a sample excess kurtosis of -1.2000 and a raw kurtosis of 1.8000. Negative excess means lighter tails than a normal distribution, which fits a roughly flat, bounded sequence. The sample and population values are close because the data are evenly spread.

Example 3: One extreme outlier

Eight zeros and one 100 produce a sample excess kurtosis of 9.0000 and a raw kurtosis of 12.0000. The outlier dominates the fourth moment, which is why kurtosis is a tail statistic. The mean rises to 11.11 and the standard deviation to 31.43, but the kurtosis is the clearest signal that the distribution is not normal.

What the result means

Excess kurtosis is best read alongside a histogram or a normal probability plot, not on its own. A positive value says the tails are heavier than normal, which means extreme observations are more likely than a normal model would suggest. A negative value says the tails are lighter and the distribution is more bounded. The sample formula is biased downward at small n without the correction, so use the sample option when the data are a sample. Even with the correction, kurtosis is noisy in small samples: a value of 0.94 from eight observations is not strong evidence of a heavy-tailed population. Use the statistic as a descriptive flag, not a hypothesis test.

Common mistakes

  • Reporting raw kurtosis as excess kurtosis, or the other way round.
  • Using the population formula on a small sample without noting the bias.
  • Treating a single outlier as proof of a heavy-tailed distribution.
  • Comparing kurtosis across data sets with very different sample sizes.
  • Ignoring the mean and variance, which are needed to interpret the standardised moment.
  • Assuming a positive excess kurtosis means the distribution is skewed; kurtosis and skewness are different.

Assumptions and limits

  • The values are a list of observations on a continuous scale.
  • The sample option treats the data as a sample from a larger population.
  • The population option treats the list as the entire population.
  • The fourth moment exists and is finite.
  • No value has been trimmed or winsorised without disclosure.
  • The result is descriptive, not a test of normality.

Frequently asked questions

What does a kurtosis of 3 mean?

Raw kurtosis of 3 is the value for a normal distribution, so excess kurtosis is 0. Values above 3 (positive excess) indicate heavier tails; values below 3 (negative excess) indicate lighter tails.

Should I use sample or population kurtosis?

Use sample when the values are a sample from a larger population, which is the usual case in research and quality control. Use population only when the list is the complete population. The sample formula corrects for bias at small n.

Why is my kurtosis so high?

A single extreme value can dominate the fourth moment. Check the data for outliers, data-entry errors, or a genuinely heavy-tailed process. A high kurtosis is a signal to look at the distribution, not a conclusion on its own.

Does kurtosis measure peakedness?

It is often described as peakedness, but it is more accurately a measure of tail weight. A distribution can have a sharp peak and light tails, or a flat peak and heavy tails. The tail interpretation is the more useful one.

How many values do I need for a stable estimate?

Kurtosis is a fourth moment and converges slowly. Small samples give unstable estimates. Use at least several dozen values for a descriptive figure, and more if the tails are the focus. The calculator will still compute with four, but treat the result as indicative.

Is kurtosis the same as skewness?

No. Skewness measures asymmetry; kurtosis measures tail weight. A symmetric distribution can have high kurtosis, and a skewed distribution can have low kurtosis. Report both if shape matters.

Sources

This is an educational statistical calculation, not a research or quality-control conclusion. Kurtosis estimates are sensitive to sample size and outliers. Use appropriate statistical software and methods for formal analysis.

2026 update

How to use the professional output

This page is not just a formula box. The result is designed to be checked, compared, and taken away.

  • Professional worksheet. The result includes the inputs used, the method and scope, and a one-variable sensitivity check for finance calculators.
  • A/B scenarios. Save two input sets in the browser and compare the primary, secondary, and breakdown figures side by side.
  • Print / PDF and CSV. Print a branded worksheet, or export the year-by-year table when the calculator produces one.

The formulas and assumptions above still govern the number. The worksheet makes the estimate auditable; it does not make it a lender quote, tax return, or personalized advice.

Calccc Editorial Team · Updated 2026-11-19

Prepared by Calccc Editorial Team as an educational estimate. This page is not a licensed professional review and is not financial, medical, legal, or tax advice.

Editorial estimate prepared by the Calccc Editorial Team; not a licensed professional review.