0
$\begingroup$

$X_1, \ldots, X_n$ iid $\sim N(0, \sigma^2)$. I want to deduce the chi-square test

$H_0 : \sigma^2 = \sigma_0^2, ~ H_1 : \sigma^2 \neq \sigma_0^2$: Decline $H_0$, if $ \displaystyle \frac{(n-1) \hat{S_n}^2}{\sigma_0^2} > \chi_{n-1, 1- \frac{\alpha}{2}}^2$ or $< \chi_{n-1, \frac{\alpha}{2}}^2$

as a likelihood ratio test. I've calculated the likelihood quotient

$\lambda(x) = \displaystyle \left( \frac{\sum_{i=1}^n x_i^2}{n \sigma_0^2} \right) ^{ \displaystyle \frac{n}{2}} \text{exp} \left(-\frac{1}{2\sigma_0^2} \sum_{i=1}^n x_i^2 + \frac{n}{2} \right)$,

but I can't see how to transformate it to get the result I want.

Can anyone help?

$\endgroup$
0
$\begingroup$

You know that $$\lambda(x)=\frac{sup_{\Theta_0}L(\theta_0|x)}{sup_{\Theta}L(\theta|x)}$$ and $$-2\log \lambda(x)\rightarrow^{Distribution}_{n\rightarrow\infty} \chi^2_1$$

$\endgroup$

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.