Why variance in R is different?

Question

Calculate variance:

v=c(1,2,3)  
print(var(v)) 
#Result:1

#Calculate manually:2/3

4

Sample vs population variance – Hugh Dec 07 '18 at 06:05
2

R uses Sample variance – Hunaidkhan Dec 07 '18 at 06:13

mickey · Answer 1 · 2018-12-07T06:38:26.337

3

There are typically two ways to estimate variance. One is 1/n * sum( (x-mean(x))^2 ) and another is 1/(n-1) * sum( (x - mean(x))^2 ). The second is an unbiased estimator for the variance, which R uses.

The comments about sample variance versus population variance seem a bit misleading to me. The population variance is usually thought of as the fixed, unknown variance of a population. We want to estimate that variance, and we do so with a sample of the data, hence a sample variance (which can be estimated in more than one way, as mentioned above).

edited Dec 07 '18 at 06:38

answered Dec 07 '18 at 06:31

mickey

2,168
2
11
20

2

Your answer is correct but you should keep in mind that finite populations exist and for some of them you can measure every member of the population. – Roland Dec 07 '18 at 07:00

score 0 · Answer 2 · answered Dec 07 '18 at 06:07

0

var in R computes sample variance. The other way of doing it is for population variance. http://www.r-tutor.com/elementary-statistics/numerical-measures/variance

answered Dec 07 '18 at 06:07

Erin

386
1
7

Why variance in R is different?

2 Answers2