Skip to content

Instantly share code, notes, and snippets.

@devries
Created April 29, 2014 16:19
Show Gist options
  • Save devries/11405101 to your computer and use it in GitHub Desktop.
Save devries/11405101 to your computer and use it in GitHub Desktop.
An implementation of the Kolmogorov-Smirnov test in python
#!/usr/bin/env python
import math
import sys
def main(argv=None):
if argv is None:
argv = sys.argv
if len(argv) != 3:
print >>sys.stderr, "Usage:",sys.argv[0],"<datafile1> <datafile2>"
sys.exit(1)
print "The KS test measures the maximum deviation (D) in the cumulative"
print "probabilities. Then the probability that the data sets are drawn"
print "from the same distribution can be calculated. Small values"
print "of the probability indicate that the distrubution of sets one"
print "and two are significantly different."
f = open(sys.argv[1],'r')
d1 = []
for ln in f:
try:
d1.append(float(ln))
except:
pass
f.close()
f = open(sys.argv[2],'r')
d2 = []
for ln in f:
try:
d2.append(float(ln))
except:
pass
f.close()
(d,prob,ne) = kstest(d1,d2)
print "D =",d
print "Prob =",prob
print "Ne =",ne
return 0
def kstest(datalist1, datalist2):
n1 = len(datalist1)
n2 = len(datalist2)
datalist1.sort()
datalist2.sort()
j1 = 0
j2 = 0
d = 0.0
fn1=0.0
fn2=0.0
while j1<n1 and j2<n2:
d1 = datalist1[j1]
d2 = datalist2[j2]
if d1 <= d2:
fn1 = (float(j1)+1.0)/float(n1)
j1+=1
if d2 <= d1:
fn2 = (float(j2)+1.0)/float(n2)
j2+=1
dtemp = math.fabs(fn2-fn1)
if dtemp>d:
d=dtemp
ne = float(n1*n2)/float(n1+n2)
nesq = math.sqrt(ne)
prob = ksprob((nesq+0.12+0.11/nesq)*d)
return d,prob,ne
def ksprob(alam):
fac = 2.0
sum = 0.0
termbf = 0.0
a2 = -2.0*alam*alam
for j in range(1,101):
term = fac*math.exp(a2*j*j)
sum += term
if math.fabs(term) <= 0.001*termbf or math.fabs(term) <= 1.0e-8*sum:
return sum
fac = -fac
termbf = math.fabs(term)
return 1.0
if __name__ == "__main__":
sys.exit(main())
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment