Have a Question?
Phone: +1 (888) 4279486
+1 (312) 2573777
Contact Us
TEST_XKURT
Calculates the pvalue of the statistical test for the population excess kurtosis (4th moment).
Syntax
TEST_XKURT^{i}(X)
X
is the data sample (a one dimensional array of cells (e.g. rows or columns)).
Remarks
 The data sample may include missing values (e.g. #N/A).

The test hypothesis for the population excess kurtosis:
Where:
 is the null hypothesis^{i}.
 is the alternate hypothesis.

For the case in which the underlying population distribution is normal, the sample excess kurtosis also has a normal sampling distribution:
Where:
 is the sample excess kurtosis (i.e. 4th moment).
 is the number of nonmissing values in the data sample.
 is the normal (i.e. Gaussian) probability distribution function.

Using a given data sample, the sample excess kurtosis is calculated as:
Where:
 is the sample excess kurtosis.
 is the ith nonmissing value in the data sample.
 is the number of nonmissing values in the data sample.
 is the sample standard deviation.
 The underlying population distribution is assumed normal (Gaussian).
 This is a twosides (i.e. twotails) test, so the computed pvalue should be compared with half of the significance level ().
Examples
Example 1:
A  B  

1  Date  Data 
2  1/1/2008  #N/A 
3  1/2/2008  2.83 
4  1/3/2008  0.95 
5  1/4/2008  0.88 
6  1/5/2008  1.21 
7  1/6/2008  1.67 
8  1/7/2008  0.83 
9  1/8/2008  0.27 
10  1/9/2008  1.36 
11  1/10/2008  0.34 
12  1/11/2008  0.48 
13  1/12/2008  2.83 
14  1/13/2008  0.95 
15  1/14/2008  0.88 
16  1/15/2008  1.21 
17  1/16/2008  1.67 
18  1/17/2008  2.99 
19  1/18/2008  1.24 
20  1/19/2008  0.64 
Formula  Description (Result)  

=KURT($B$2:$B$20)  Sample excess kurtosis (1.0517)  
=TEST_XKURT($B$2:$B$20)  pvalue of the test when excess kurtosis = 0 (0.171) 