def correlation(sd, a, b):
sampleA = SelectWithSeed.pickItem(sd, a, 10)
sampleB = SelectWithSeed.pickItem(sd, b, 10)
cov = Covariance.covariance(sampleA, sampleB)
stdDevA = StandardDeviation.standardDeviation(sampleA)
stdDevB = StandardDeviation.standardDeviation(sampleB)
return cov / (stdDevA * stdDevB)
def popcorrelation(data1, data2):
covxy = 0
n = 0
xmean = Mean.mean(data1)
ymean = Mean.mean(data2)
xstd = StandardDeviation.stardardDev(data1)
ystd = StandardDeviation.stardardDev(data2)
if (len(data1) == len(data2)):
for x, y in zip(data1, data2):
n += 1
covxy += abs((int(x - xmean) * int(y - ymean)) / n)
return covxy / (xstd * ystd)
def samplecorrelation(data1, data2):
numerator = 0
n = 0
meanofx = Mean.mean(data1)
meanofy = Mean.mean(data2)
stdevx = StandardDeviation.stardardDev(data1)
stdevy = StandardDeviation.stardardDev(data2)
if len(data1) == len(data2):
for x, y in zip(data1, data2):
n += 1
numerator += abs(
(int(x - meanofx) * int(y - meanofy)) / (n - 1))
return numerator / (stdevx * stdevy)
def setUp(self):
self.sample = Sample()
self.list1 = Randm.randList(1, 100, 30, 99)
self.size = len(self.list1)
self.standarddev = StandardDeviation.stardardDev(self.list1)
self.merror = 3
self.mean = Mean.mean(self.list1)
def confInterval(data, cval=0.95):
mean = Mean.mean(data)
standardDev = StandardDeviation.stardardDev(data)
n = len(data)
zscr = ZScore.zscore(cval)
num = round(zscr * (standardDev / nthroot.root(2, n)), 2)
return round((mean - num), 2), round(mean), round((mean + num), 2)
def marginoferror(data):
standardDev = StandardDeviation.stardardDev(data)
zscr = abs(ZScore.zscore(data))
n = len(data)
return zscr * (standardDev / (nthroot.root(2, n)))
def confInterval(data):
zvalue = ZScore.zscore(data)
stdDev = StandardDeviation.stardardDev(data)
mean = Mean.mean(data)
n = len(data)
num = round(zvalue * (stdDev / nthroot.root(2, n)), 2)
return round((mean - num), 2), round(mean), round((mean + num), 2)
def sampleSize(sd, data):
e = MarginError.margin(sd, data)
stdDev = StandardDeviation.standardDeviation(data)
val = (1.96 * stdDev) / e
sample = Exponentiation.exponent(val, 2)
return sample
def zscore(data):
newData = []
mean = Mean.mean(data)
stdDev = StandardDeviation.stardardDev(data)
for i in data:
newData.append((i - mean)/stdDev)
return newData
def test_StatisticFunctions_StandardDeviation(self):
self.assertEqual(29.052495589880053, StandardDeviation.standardDeviation(self.testData))
def variance(data):
return Exponentiation.power(StandardDeviation.stardardDev(data), 2)
def standardDev(self, data):
self.result = StandardDeviation.stardardDev(data)
return self.result
def samplestdev(data):
zval = ZScore.zscore(data)
stdDev = StandardDeviation.stardardDev(data)
merror = MarginOfError.marginoferror(data)
num = Exponentiation.power(((zval * stdDev) / (merror / 100)), 2)
return round(num, 2)
def popCor(a, b):
cov = Covariance.covariance(a, b)
stdDevA = StandardDeviation.standardDeviation(a)
stdDevB = StandardDeviation.standardDeviation(b)
return cov / (stdDevA * stdDevB)
def skewness(data):
mean = Mean.mean(data)
median = Median.med(data)
stdDev = StandardDeviation.stardardDev(data)
return (3 * (mean - median)) / stdDev
def margin(sd, data):
z_score = Zscore.zscore(sd, data)
stdDev = StandardDeviation.standardDeviation(data)
return z_score * stdDev