What it means
Statisticians describe the shape of data using several measures, and kurtosis tells us about the tails. A normal distribution has a kurtosis of 3, and distributions are compared with that benchmark.
If kurtosis is below 3, the distribution is platykurtic; if it is close to 3, it is mesokurtic; and if it is above 3, it is leptokurtic. Many people subtract 3 to give excess kurtosis, so that the normal distribution has a value of 0.
A platykurtic distribution then has negative excess kurtosis. Software often reports excess kurtosis by default, so it is important to check which version is being used.
The word comes from Greek roots meaning broad or flat, which describes the lower peak and the wide spread across the middle. The data is spread more evenly, and there are fewer extreme values far from the average.
A uniform distribution, where every outcome is equally likely, is an example. In finance, most investment returns are not platykurtic.
Market returns tend to be leptokurtic, with fat tails, meaning that big losses and gains happen more often than a normal curve implies. A platykurtic series would be unusually stable, and finding one may suggest that smoothing, limited data or a managed process is involved.
Risk managers care because tail events drive losses. If they assume a normal curve when the data is leptokurtic, they underestimate extreme outcomes, and if the data is platykurtic, the normal curve overstates them.
Checking kurtosis helps them choose the right model and avoid false comfort. Business users meet the idea in quality control, forecasting and budgeting.
A process with tightly bounded results, such as a machine that fills bottles within a narrow range, may show a platykurtic pattern. Understanding the shape helps managers set realistic limits and judge how much to trust a single average figure.
In practice
Real-world examples.
Example
A factory fills bottles with a machine that keeps the volume within a narrow band, and the quality team plots the results. The pattern is flat on top with no extreme outliers. The team describes it as platykurtic and sets control limits based on the true spread.
Example
A fund manager reviews monthly returns that range between -2% and +2% with no unusual spikes. The kurtosis is below 3. She suspects that the returns are being smoothed and asks the administrator how the assets are valued.
Example
A retailer studies daily sales across 20 similar shops and finds they are spread evenly across a limited range. The analyst calls the data platykurtic. She builds the forecast with a range of outcomes rather than assuming a bell curve.
Formula
Calculation
Kurtosis = [sum of (x - mean)^4 / n] / (standard deviation)^4
Excess kurtosis = kurtosis - 3
Take five days of profit, in thousands of dollars: $1,000, $2,000, $3,000, $4,000 and $5,000, written as 1, 2, 3, 4 and 5. The mean is 3. The deviations are -2, -1, 0, 1 and 2, and their squares are 4, 1, 0, 1 and 4, which sum to 10.
The variance (using all n = 5 points) is 10 / 5 = 2, so (standard deviation)^4 = 2 x 2 = 4. The fourth powers of the deviations are 16, 1, 0, 1 and 16, which sum to 34, and 34 / 5 = 6.8. Kurtosis is 6.8 / 4 = 1.7, and excess kurtosis is 1.7 - 3 = -1.3, so the data is platykurtic.Case study
Seen in the real world.
Brightfield Analytics is a fictional consultancy, and this story is illustrative. A client asked it to model the daily profit of a set of vending machines, and the analyst first assumed a normal bell curve.
The data showed daily profit ranging only from $1,000 to $5,000, with each level occurring about equally often, and the kurtosis worked out at about 1.8, well below 3. Using a normal curve would have suggested occasional profits of $8,000 or losses, which never happened.
The analyst used a bounded distribution instead and gave the client a more realistic range for budgeting. The client was able to set a cash reserve at a lower level without taking extra risk. The illustrative lesson is that checking the shape of the data prevents both overestimating and underestimating extreme outcomes.
Watch out
Common mistakes.
- Thinking a platykurtic distribution has no risk, when it simply has fewer extreme outcomes than a normal curve.
- Mixing up kurtosis and excess kurtosis, which differ by 3.
- Assuming all financial returns are platykurtic, when most have fat tails.
Questions
People also ask.
What is the opposite of platykurtic?
Leptokurtic, which has a sharper peak and fatter tails than the normal curve.
What kurtosis value makes data platykurtic?
A kurtosis below 3, or an excess kurtosis below 0.
Why does it matter in finance?
Because the shape of the tails affects how likely extreme gains or losses are, and so how risk should be measured.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%Related