0
$\begingroup$

$X_1,\cdots ,X_n $ random sample from $N(\mu,\sigma^2)$

I know, that distribution of the random variable $$\frac{(n-1)S^2}{\sigma^2}$$ has a chi-square distribution with $(n-1)$ degrees of freedom, where $$S^2=\frac{1}{n-1}\sum^{n}_{i=1}(X_i-\bar{X})^2.$$

Is it true, that $\frac{1}{n}\sum_{n=1}^{m} (X_i - \mu)^2$ has $\frac{\chi^2_n \sigma^2}{n}$ distribution?

$\endgroup$
  • $\begingroup$ Neither statement is true unless $\sigma^2=1$. $\endgroup$ – whuber Jan 17 '18 at 18:53
  • $\begingroup$ @whuber sorry, I corrected. Have a look please. $\endgroup$ – Rob_Fir Jan 17 '18 at 18:57
  • 1
    $\begingroup$ The answer to your question is contained in the first line of the Wikipedia article on the chi-squared distribution. $\endgroup$ – whuber Jan 17 '18 at 19:03
1
$\begingroup$

Yes, it's true: The chi-squared distribution with $n$ degrees of freedom is defined as the distribution of a sum of squares of $n$ independent standard normal random variables $Z_i$. So the most basic defining result for chi-squared random variables is that for $X_1, ..., X_n \text{ ~ IID N}(\mu, \sigma^2)$ we have:

$$\sum_{i=1}^n \left(\frac{X_i - \mu}{\sigma}\right)^2 = \sum_{i=1}^n Z_i^2 \text{ ~ } \chi_n^2.$$

Multiplying both sides by $\sigma^2/n$ gives the result:

$$\frac{1}{n} \sum_{i=1}^n (X_i - \mu)^2 \text{ ~ } \frac{\chi_n^2 \cdot \sigma^2}{n}.$$

| cite | improve this answer | |
$\endgroup$
0
$\begingroup$

See Cochran's Theorem...

Cochran's Theorem tells when a variable follows normal distribution then square of the same Normal distributed variable will follow Chi-Square distribution in some manner. Now here, the samples are taken from the normal distribution so squaring those samples and taking sum over those squared samples will follow Chi-Square distribution in some manner. Now, sum of squaring the mean deviated samples is nothing but the sample variance $(n-1)S^2$. Now, by Cochran's theorem, $y = \frac{\sum_{i=1}^n (X_i-\bar{X})^2}{\sigma^2} \sim \chi^2_{n-1}$. So $\sum_{i=1}^n (X_i-\bar{X})^2 \sim\chi^2_{n-1} \sigma^2$.

This is my explanation.

| cite | improve this answer | |
$\endgroup$
  • 1
    $\begingroup$ We get additional degree of freedom, because we substitute $\bar X$ on $\mu$? RIght? $\endgroup$ – Rob_Fir Jan 17 '18 at 19:59

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.