Cramér-Rao inequality

If $\hat{θ}$ is an unbiased estimator of the parameter $θ$ , then the variance of $\hat{θ}$ must satisfy the inequality

v a r (\hat{θ}) \geq \frac{1}{n E [{(\frac{\partial \ln f (x; θ)}{\partial θ})}^{2}]}

CRLB $\to 0$ as $n \to \infty$