Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

So I have a problem with my homework, and I just need to see if my proof and thinking are correct. The problem I have is this: Show that the sample variance $s^2$ is a consistent estimator for the variance of X, $\sigma^2$. I had to manipulate the $\sigma^2$ shortcut formula to get $s^2=\sigma^2\frac{n}{n-1}$.

However, my professor gave us a hint to use the theorem of consistent estimators for moments of X.

This theorem goes: Let X be a random variable with $k^{th}$ moment $\mathrm{E}[X^k]$ unknown. Then $\frac{\sum_iX_i^k}{n}$ is a consistent estimator for $\mathrm{E}[X^k]$.

So I am thinking that when I get to my proof where I have $\frac{n}{n-1}(\mathrm{E}[X^2]-\mathrm{E}[X]^2)$, is where I can say that by this theorem that since $\mathrm{E}[X^2]$ and $\mathrm{E}[X]$ are consistent estimators for $\mathrm{E}[X^k]$ moment, that $s^2$ is a consistent estimator for $\sigma^2$.

Could anyone let me know if my thinking is correct on this? Sorry for my code not being so clean, this is my first time really programming in LaTeX. Thanks!

share|cite|improve this question

I guess $s^2$ is the unbiased estimator for the variance $\frac{1}{n-1}\sum_{i=1}^n (x_i-\bar x)$. It is consistent, if it converges in probability to $\sigma^2$, which means that $\lim_{n\to\infty}P(|s^2-\sigma^2|\geq\epsilon)=0$ for any $\epsilon>0$. Now you know by your Professor's theorem that the estimator $s_n^2=\frac{1}{n}\sum_{i=1}^n (x_i-\bar x)$ is consistent, i.e. $\lim_{n\to\infty}P(|s_n^2-\sigma^2|\geq\epsilon)=0$. Using that it shouldn't be too hard to show, that the other limit also goes to zero.

share|cite|improve this answer
I understand that it is consistent in that regard, but should I just take the limits of each of the expected values as they go to infinity just to show that this formula converges in probability to 0? – Perdue Feb 28 '13 at 17:58

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.