def testVariance(self):
data = [1, 2, 3]
assert stats.mean(data) == 2
self.assertEqual(stats.pvariance(data), 2/3)
self.assertEqual(stats.variance(data), 1.0)
self.assertEqual(stats.pstdev(data), math.sqrt(2/3))
self.assertEqual(stats.stdev(data), 1.0)
def testManyDuplicates(self):
from stats import pvariance
# Start with 1000 normally distributed data points.
data = [random.gauss(7.5, 5.5) for _ in range(1000)]
expected = pvariance(data)
# We expect a to be close to the exact result for the variance,
# namely 5.5**2, but because it's random, it might not be.
# Either way, it doesn't matter.
# Duplicating the data points should keep the variance the same.
for n in (3, 5, 10, 20, 30):
d = data*n
actual = pvariance(d)
self.assertApproxEqual(actual, expected, tol=1e-12)
# Now try again with a lot of duplicates.
def big_data():
for _ in range(500):
for x in data:
yield x
actual = pvariance(big_data())
self.assertApproxEqual(actual, expected, tol=1e-12)
def testPVar(self):
result = stats.pvariance(self.data)
expected = self.expected["pvariance"]
self.assertAlmostEqual(result, expected, places=self.places)
def testPVar(self):
sx2 = stats.pvariance(self.xdata)
sy2 = stats.pvariance(self.ydata)
self.assertAlmostEqual(sx2, 11004687 / 1048576, places=self.places)
self.assertAlmostEqual(sy2, 4647 / 4096, places=self.places)
