Give a sample $X_1,X_2,\ldots,X_n $~$N(\mu,\sigma^2)$ and $\sigma^2=1$. I want to find Fisher's information. I know that: $$ -E(\frac{d^2}{d\mu^2} \ln f(x))=1/\sigma^2. $$ Why this is the Fisher's information in this case? Why the second derivative of normal distribution is not simply equal to zero as expectation of $X$ is $\mu$?

  • If the parameter is $\mu$, $\ln f(x;\mu)=const-\frac{(x-\mu)^2}{2\sigma^2}$. The second derivation is not zero but simply $-\frac{1}{\sigma^2}$. – Arash Sep 1 '17 at 10:10
  • Yes, we treat $\mu$ as parameter. OK but if we expand the top of the fraction after derivative we are left with $X-\mu$ which should be equal to 0? Could you explain that further? – Mrowkacala Sep 1 '17 at 10:22
  • Take the second derivative of $-\frac{(x-\mu)^2}{2\sigma^2}$; how do you end up with $X-\mu$? – Arash Sep 1 '17 at 10:25
  • It shouldn't be a second derivative of normal distribution PDF? If not than i understand the reason why however i thought that it is about second derivative of PDF function. – Mrowkacala Sep 1 '17 at 10:26
  • 1
    its logarithm: $\ln f(x)$ – Arash Sep 1 '17 at 10:27

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.