I have trouble solving following problem:

Let $$X_1,...,X_n$$ be independently,identically distributed random variables with

the Riemann-density $$f$$ on a probability space $$(\Omega,\mathcal{A},P).$$

Moreover denote $$\sigma^2=Var(X_1)$$ and let $$\int_{\Bbb R}x^2 f(x)dx<\infty.$$

We have to show that:


I would really appreciate any help and I hope everything is correctly written.



First assume that $E[X_i]=0$, then expand out the square. You'll need to be able to calculate $E[X_i X_j]$ for $i=j$ as well as $i \neq j$. To handle the case $E[X_i]=\mu \neq 0$, consider

$$X_i - \frac{1}{n} \sum_{j=1}^n X_j = (X_i - \mu) - \frac{1}{n} \sum_{j=1}^n (X_j-\mu)$$

which reduces the problem to the previous case.


Let $\mu:=\mathbb{E}X_{1}$ and for convenience assume that $\mu=0$ and $\sigma^{2}=1$.


Here $\mathbb{E}X_{i}X_{j}=\mathsf{Cov}\left(X_{i},X_{j}\right)=0$ if $i\neq j$ and $\mathbb{E}X_{i}^{2}=\mathsf{Var}X_{i}=1$.

So applying linearity of expectation we find:

$\mathbb{E}\left(X_{1}-\overline{X}\right)^{2}=1-\frac{2}{n}+\frac{1}{n^{2}}n=1-\frac{1}{n}$ so that $\mathbb{E}\frac{1}{n-1}\sum_{i=1}^{n}\left(X_{i}-\overline{X}\right)^{2}=\mathbb{E}\frac{1}{n-1}n\left(1-\frac{1}{n}\right)=1$.

In the more general case it can be exploited that we can write $X_{i}=\sigma U_{i}+\mu$ where $U_{i}$ has mean $0$ and variance $1$.

This leads easily to: $\sum_{i=1}^{n}\left(X_{i}-\overline{X}\right)^{2}=\sigma^{2}\sum_{i=1}^{n}\left(U_{i}-\overline{U}\right)^{2}$.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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