Updated: 29 June 2012

Use KSccdf to compute the complementary cumulative probability P[D

SELECT [wctStatistics].[wct].[KSccdf](

<@n, int,>

,<@k, float,>)

is the sample size. @*n *is an expression of type **int** or of a type that can be implicitly converted to **int**.

is the maximum magnitude of the discrepancy between the empirical and proposed distributions. @*x *is an expression of type **float** or of a type that can be implicitly converted to **float**.

Float

· If @n < 0 then KSccdf = 1

· If @x < 0 then KSccdf = 1

SELECT n

,sqrt(18.0/n) as x

,wct.KSCCDF(n, sqrt(18.0/n)) as ks

FROM (VALUES (50.0),(100.0),(500.0),(1000.0),(5000.0),(10000.0),(50000.0),(10E5),(10E6),(10E7),(10E8))p(n)

This produces the following result.

n x ks

---------------------- ---------------------- ----------------------

50 0.6 9.6340704561422E-18

100 0.424264068711929 7.60653219848608E-17

500 0.189736659610103 3.09340954271647E-16

1000 0.134164078649987 3.69599264243387E-16

5000 0.06 4.33712332368773E-16

10000 0.0424264068711929 4.44862619998983E-16

50000 0.0189736659610103 4.56828378005821E-16

1000000 0.00424264068711928 4.62531376662394E-16

10000000 0.00134164078649987 4.63483518913332E-16

100000000 0.000424264068711929 4.63772741835367E-16

1000000000 0.000134164078649987 4.63863011932784E-16