module KolmogorovSmirnov: sig
.. end
val goodness_of_fit : float array ->
cumulative_probability:(float -> float) ->
?alternative:Tests.test_alternative -> unit -> Tests.test_result
One-sample Kolmogorov-Smirnov test for goodness of fit, which
evaluates the distribution G(x)
of the observed random variable
against a given distribution F(x)
. Under the null hypothesis
the two distributions are identical, G(x) = F(x)
.
val two_sample : float array ->
float array ->
?alternative:Tests.test_alternative -> unit -> Tests.test_result
Two-sample Kolmogorov-Smirnov test, which evaluates the null
hypothesis, that two
independent samples are drawn from the
same continious distribution.
Note: in the current implementation samples with ties will
result in an Invalid_argument
exception.
References
- National Institute of Standards and Technology (US), et al.
"Engineering statistics handbook", Section 1.3.5.16.
The Institute, 2001.
- Jingbo Wang, Wai Wan Tsang, and George Marsaglia.
"Evaluating Kolmogorov's distribution." Journal of
Statistical Software 8, no. 18. 2003.
- Z. W. Birnbaum, Fred H. Tingey. "One-sided confidence contours
for probability distribution functions." The Annals of
Mathematical Statistics, pp592-596. 1951.