Kstest: Difference between revisions

From Eigenvector Research Documentation Wiki
Jump to navigation Jump to search
imported>Jeremy
No edit summary
imported>Jeremy
No edit summary
 
Line 6: Line 6:


:vals = kstest(x,distname)
:vals = kstest(x,distname)
===Description===
Performs a Kolmogorov-Smirnov test on a set of values to determine if they came from a specified distribution.


====Inputs====
====Inputs====
Line 15: Line 19:
====Outputs====
====Outputs====


The return value is a structure with fields (larger values indicate rejecting the named distribution as a candidate parent distribution for the sample). The ks is the value of the Kolmogorov-Smirnov statistic and is   times the maximum difference of the distributions. The maximum difference in the distributions is returned as Dn.
The return value is a structure with fields (larger values indicate rejecting the named distribution as a candidate parent distribution for the sample). The ks is the value of the Kolmogorov-Smirnov statistic and is <math>\sqrt[ ]{ n }</math> times the maximum difference of the distributions. The maximum difference in the distributions is returned as Dn.


* '''Ks''' =  value of the adjusted test statistic.
* '''Ks''' =  value of the adjusted test statistic.

Latest revision as of 08:18, 4 September 2008

Purpose

Kolmogorov-Smirnov test that a sample has a specified distribution.

Synopsis

vals = kstest(x,distname)

Description

Performs a Kolmogorov-Smirnov test on a set of values to determine if they came from a specified distribution.

Inputs

  • x = matrix (column vector) in which the sample data is stored.
  • distname = string, optional distribution name to assume as the parent distribution for the sample. Default value is 'normal'.

Outputs

The return value is a structure with fields (larger values indicate rejecting the named distribution as a candidate parent distribution for the sample). The ks is the value of the Kolmogorov-Smirnov statistic and is times the maximum difference of the distributions. The maximum difference in the distributions is returned as Dn.

  • Ks = value of the adjusted test statistic.
  • Dn = unadjusted test statistic.
  • parameters = maximum likelihood estimates.

Examples

 kstest(x)
 kstest(x,'exp')

See Also

chitest, distfit