KSTEST2_q
Updated: 27 June 2012
Use the scalra function KSTEST2_q to perform the two-sample Kolmogorov-Smirnov test to compare the distributions of the values in two samples. KSTEST2 produces the test statistic (k) and the p-value (p) which can be used to ascertain if the two samples are from the same continuous distribution.
Syntax
SELECT [wctStatistics].[wct].[KSTEST2_q](
<@X_RangeQuery, nvarchar(max),>
,<@Statistic, nvarchar(4000),>)
Arguments
@X_RangeQuery
a string which contains the SQL to select the sample values and the cumulative distiribution function. The resultant table must contain two columns containing the following values.
@sid
the sample identifier. @sid indentifies which of the two samples @ x belongs to.
@x
the sample value. @x is an expression of type float or of a type that can be implicitly converted to float.
@Statistic
identifies the return value as either the Kolmogorov-Smirnov statistic (k) or the p-value (p).
Return Types
float
Remarks
· If @x is null then @x is not included in the calculation.
· There must exactly 2 @sid in the resultant table from @X_RangeQuery.
· To evaluate one sample against a continuous distribution function, use KSTEST or KSTEST_q
· For simpler queries, consider using KSTEST2
Examples
In this example, we compare a sample of test scores for men and women
/*Put data into a temp table*/
SELECT *
INTO #n
FROM (
SELECT 'Men',117 UNION ALL
SELECT 'Men',106 UNION ALL
SELECT 'Men',105 UNION ALL
SELECT 'Men',112 UNION ALL
SELECT 'Men',108 UNION ALL
SELECT 'Men',115 UNION ALL
SELECT 'Men',97 UNION ALL
SELECT 'Men',112 UNION ALL
SELECT 'Men',110 UNION ALL
SELECT 'Women',131 UNION ALL
SELECT 'Women',107 UNION ALL
SELECT 'Women',110 UNION ALL
SELECT 'Women',108 UNION ALL
SELECT 'Women',100 UNION ALL
SELECT 'Women',74 UNION ALL
SELECT 'Women',89 UNION ALL
SELECT 'Women',115 UNION ALL
SELECT 'Women',85 UNION ALL
SELECT 'Women',94 UNION ALL
SELECT 'Women',110 UNION ALL
SELECT 'Women',80 UNION ALL
SELECT 'Women',103 UNION ALL
SELECT 'Women',122
) n(s,x)
/*Calculate the test statistics and the p-value*/
SELECT wct.KSTEST2_q('SELECT s,x from #n','p') as p_value
,wct.KSTEST2_q('SELECT s,x from #n','k') as k_observed
/*Clean up the temp table*/
DROP TABLE #n
This produces the following result
p_value k_observed
---------------------- ----------------------
0.297578744560361 0.388888888888889