Buzz
The article below introduces you to the KURT function - one of the functions in the statistical group widely used in Excel.
Description: The function returns the kurtosis coefficient of the dataset. Kurtosis coefficient represents the peakedness or flatness of the dataset.
- If the kurtosis coefficient is positive -> relatively peaked distribution.
- If the kurtosis coefficient is negative -> relatively flat distribution.
Syntax: KURT(number1, [number2], ...)
Where:
- number1, [number2], ...): The argument for which to calculate the kurtosis coefficient, where number1 is required, and the rest are optional and can contain up to 255 number arguments.
Note:
- The values of the arguments must be numbers, names, arrays, or references containing numbers.
- Logical or text values entered directly into the argument list -> these values are still calculated.
- If the argument is an array reference containing text or logical values -> these values are ignored, however, the value 0 is still calculated.
- If the kurtosis coefficient is negative -> relatively flat distribution.
- If the number of data points is less than 4 or the standard deviation is 0 -> the function returns an error value #DIV/0
- Kurtosis is determined by the formula:
Where: s represents the sample standard deviation.
Example:
Find the kurtosis coefficient of the dataset in the table below:
- In the cell where you want to calculate, enter the formula: =KURT(D6:D10)
- Press Enter -> the kurtosis coefficient of the dataset is:
The kurtosis coefficient of the dataset above is less than 0 -> indicating a flat distribution.
- In cases where the number of data points in two arrays x, y are different and include logical or text values -> the function disregards these values.
- In cases where the number of data points in the dataset is less than 4 -> the function returns an error value #DIV/0
Above is the guide and some specific examples when utilizing the KURT function in Excel.
Best wishes for your success!
