0
$\begingroup$

Let $X_1,\dots,X_n$ be a random sample from the density

$f(\theta;x)=\theta x^{-2}I_{[\theta,\infty)}(x)$

where $\theta>0$. Is $X_{(1)}=\min\left[X_i\right]$ a sufficient statistic?

My attempt

I made the joint density function like so:

\begin{align} L\left(\theta; x_1,\dots, x_n \right)&=\prod_{i=1}^n\theta x_i^{-2}I_{[\theta,\infty)}(x_i) \\ &= \theta^n\prod_{i=1}^nx_i^{-2}\prod_{i=1}^nI_{[\theta,\infty)}(x_i) .\end{align}

Now I want to rewrite $\prod_{i=1}^nI_{[\theta,\infty)}(x_i)$ in terms of $X_{(1)}$, but I'm not sure how to do. Any suggestions?

$\endgroup$
0
$\begingroup$

Observe that $$\prod_{i=1}^n \mathbb 1(x_i \ge \theta) = \mathbb 1\left( \bigcap_{i=1}^n x_i \ge \theta \right) = \mathbb 1 \left(\min_i x_i \ge \theta\right) = \mathbb 1 (x_{(1)} \ge \theta).$$

$\endgroup$

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.