Performing the KS-test on a scaled distribution in python

Question

I want to perform the KS-test on my data by comparing it to a uniform distribution that has been scaled.

How do I pass just a scale parameter to the args in scipy.stats.kstest (without passing any other argument to it)? I want it to take the default value for the other arguments (egs:loc)

See:
1. https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.uniform.html
2. https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.kstest.html

I tried the following but it did not work (invalid syntax)

stats.kstest(data, 'uniform',args=(scale=15.0))

Answer 1

In this case, create a lambda for the distribution you want and divide the x parameter by the desired scale. An example follows:

import random
import scipy.stats as stats
data=[random.random()*15 for i in range(10000)]
stats.kstest(data, lambda x: stats.uniform.cdf(x/15.0))

An alternative approach, which works for the distributions found in SciPy:

stats.kstest(data, lambda x: stats.uniform.cdf(x,scale=15.0))