TATTER: Two-sAmple TesT EstimatoR
Abstract
TATTER (Two-sAmple TesT EstimatoR) performs two-sample hypothesis test. The two-sample hypothesis test is concerned with whether distributions p(x) and q(x) are different on the basis of finite samples drawn from each of them. This ubiquitous problem appears in a legion of applications, ranging from data mining to data analysis and inference. This implementation can perform the Kolmogorov-Smirnov test (for one-dimensional data only), Kullback-Leibler divergence, and Maximum Mean Discrepancy (MMD) test. The module performs a bootstrap algorithm to estimate the null distribution and compute p-value.
- Publication:
-
Astrophysics Source Code Library
- Pub Date:
- June 2020
- Bibcode:
- 2020ascl.soft06007F
- Keywords:
-
- Software