Kurtosis: Difference between revisions
Jump to navigation
Jump to search
imported>Scott (Created page with "===Purpose=== Returns the kurtosis statistic for a vector or matrix. ===Synopsis=== : k = kurtosis(x) : k = kurtosis(x,flag,dim) ===Description=== Returns the kurtosis st...") |
imported>Scott No edit summary |
||
| Line 24: | Line 24: | ||
====Outputs==== | ====Outputs==== | ||
* '''k''' = kurtosis of x over the specified dimension (dim). | * '''k''' = kurtosis of x over the specified dimension (dim). | ||
===See Also=== | |||
[[skewness]]] | |||
Revision as of 19:13, 8 June 2015
Purpose
Returns the kurtosis statistic for a vector or matrix.
Synopsis
- k = kurtosis(x)
- k = kurtosis(x,flag,dim)
Description
Returns the kurtosis statistic for a vector or matrix, x. Missing data is ignored for the calculation.
Inputs
- x = Vector or matrix.
Optional Inputs
- flag = bias-correction disable flag. When set to 0 (zero), the kurtosis is corrected for sampling bias. Default is 1 (one) to return the uncorrected kurtosis. The correction for bias is done by the formula:
- k = (n-1)/((n-2)*(n-3)) * ((n+1)*k-3*(n-1))+3
- where n is the number of non-missing elements in the given column of x.
- dim = dimension of x over which to calculate the kurtosis. Default is 1, over rows (kurtosis calculated down columns.)
Outputs
- k = kurtosis of x over the specified dimension (dim).