The Ultimate Algorithms for calculating variance - American History Information Guide and Reference

In statistics, a formula for calculating the variance of a population of size n is:

$\mathrm{Variance} = \frac {n\sum_{i=1}^{n} x_i^2 - (\sum_{i=1}^{n} x_i)^2}{n^2}.$

A formula for calculating the unbiased estimation of the population variance from n finite samples is:

$\mathrm{Variance} = \frac {n\sum_{i=1}^{n} x_i^2 - (\sum_{i=1}^{n} x_i)^2}{n(n-1)}.$

The method of calculation may be more easily understood from the table below where the mean is 8.

 i   xi   xi − mean 
 (xi − mean)2

 (index)  (datum)   (deviation)  (squared deviation)

 1  5   −3  9

 2  7   −1  1

 3  8   0  0

 4  10   2  4

 5  10   2  4

 n = 5  sum = 40   0  18

mean = 40/5 = 8
variance = (5 · 338 − 40²)/(5 · 4) = 4.5
standard deviation = $\sqrt{\mathrm{Variance}} = 2.12$

Note: Details of the variance calculation:

338 = [5² + 7² + 8² + 10² + 10²]
40 = [5 + 7 + 8 + 10 + 10]

Algorithm

Therefore a simple algorithm to calculate variance can be described by the following pseudocode:

long n = 0;
double sum = 0;
double sum_sqr = 0;
double variance;

foreach x in data:
  n += 1;
  sum += x;
  sum_sqr += x*x;
end for

variance = (sum_sqr - sum*sum/n)/(n-1);

Algorithm

Another algorithm which avoids large numbers in sum_sqr while summing up

double avg = 0;
double var = 0;
long n = data.length; // number of elements

for i = 1 to n
 avg = (avg*i + data[i]) / (i + 1);
 var = (var * (i - 1) + (data[i] - avg)*(data[i] - avg)) / i;
end for

return var; // resulting variance

Categories: Statistics

Your American History Reference Guide! - Algorithms for calculating variance

Algorithms for calculating variance

Algorithm

Algorithm

Your American History Reference Guide!
- Algorithms for calculating variance