Last active
July 27, 2017 13:00
-
-
Save MarkusPrim/206bd81f388aa9f224016b48f99dec50 to your computer and use it in GitHub Desktop.
Cramer-von-Mises distance for two weighted samples
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import numpy | |
def cvm_2samp_weigted(data1, data2, wei1, wei2): | |
"""Calculate Cramer-vonMises test statistics with weighted samples.""" | |
# Indices for a sorted array | |
ix1 = numpy.argsort(data1) | |
ix2 = numpy.argsort(data2) | |
# Sorted data | |
data1 = data1[ix1] | |
data2 = data2[ix2] | |
# Sorted weights | |
wei1 = wei1[ix1] | |
wei2 = wei2[ix2] | |
data = numpy.concatenate([data1, data2]) | |
cwei1 = numpy.hstack([0, numpy.cumsum(wei1)/sum(wei1)]) | |
cwei2 = numpy.hstack([0, numpy.cumsum(wei2)/sum(wei2)]) | |
cdf1we = cwei1[[numpy.searchsorted(data1, data, side='right')]] | |
cdf2we = cwei2[[numpy.searchsorted(data2, data, side='right')]] | |
return sum((f1-f2)**2 for f1, f2 in zip(cdf1we, cdf2we)) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment