runningStatistics.py

import math 

def running_mean(x, i, mean, delta):
    delta = x - mean
    mean  = mean + delta / i
    return mean, delta

def running_variance(x, i, mean, msq):
    delta = x - mean
    mean  = mean + delta / i
    msq   = msq + delta * ( x - mean )
    return msq, mean, delta

def calcAverage(s):
    sum = 0
    for i in range( len(s) ):
        sum = sum + s[i]
    return sum/len(s)

def calcVariance(s):
    sum = 0
    sumsqrd = 0
    for i in range( len(s) ):
        sum = sum + s[i]
        sumsqrd = sumsqrd + s[i] ** 2
    return (sumsqrd + n * (calcAverage(s)) ** 2 - 2 * (calcAverage(s)) * sum )/len(s)

if __name__ == "__main__":

    samples = [1640, 1640.5, 1640.3, 1641, 1640.7, 1640.5, 1640.7, 1640.9, 1640, 1641]
    n = len( samples )
    print("num samples: ", n)

    print("Average: ", calcAverage(samples))
    print("Variance: ", calcVariance(samples))

    # Running algorithms
    i = 0
    mean = 0
    delta = 0
    for x in samples:
        i = i + 1
        mean, delta = running_mean(x, i, mean, delta)
    print("Running Average: ", mean)

    i = 0
    mean = 0
    delta = 0
    msq = 0
    for x in samples:
        i = i + 1
        msq, mean, delta = running_variance(x, i, mean, msq)
    print("Running Variance: ", msq / n)