For the question in the link below, I am seeking the minimal sufficient statistic for $\theta$={$\beta_1$,$\beta_2$} in the linear regression model given.

enter image description here

I have taken the ratio of likelihoods $L(x, \theta)$/$L(y,\theta)$ which has given me the result below:

enter image description here

I know that I am looking for a condition that makes the likelihood ratio independent of $\theta$. From the first term in the product, when the sum of all $x_i$ is equal to the sum of all $y_i$, this term is made independent of $\theta$.

What about the second term in this product? Is there a second condition from this fraction that I need to synthesise into the solution, or have I already shown that the minimal sufficient statistic is the sum of all $x_i$? I can't find a way to algebraically break up that second fraction so that the product symbol goes away on each side.



Your Answer

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

Browse other questions tagged or ask your own question.