Verifying Central Limit Theorem computationally

central_limit_theorem.py

from random import randint
from matplotlib import pyplot

def gen_mean(sample_size):
	sum = 0
	for i in range(sample_size):
		sum += randint(1, 100)

	mean = ( float(sum) / sample_size )
	return mean

sample_means = []
sample_size = 100
no_of_samples = 100000

for c in range(no_of_samples):
	sample_means.append( int(gen_mean(sample_size)) )

min_mean = int( min(sample_means) )
max_mean = int( max(sample_means) +1 )

# Create graph 
X = range(min_mean, max_mean)
Y = [ sample_means.count(i) for i in X ] 
print X
print Y
pyplot.plot( X, Y, '-')
pyplot.xlabel( 'Sample Mean')
pyplot.ylabel( 'No of occurences of sample mean' )
pyplot.show()