= Solution
The estimated <standard error> is $\widehat\theta^{3/2}/\sqrt n$. Using the central 95% <standard normal quantiles> $\pm1.96$, a <Wald confidence interval> is
$$
\boxed{\left[\widehat\theta-1.96\frac{\widehat\theta^{3/2}}{\sqrt n},\quad\widehat\theta+1.96\frac{\widehat\theta^{3/2}}{\sqrt n}\right]\cap(0,\infty)}.
$$
Consistency of $\widehat\theta$ and <Slutsky's theorem> justify replacing the unknown scale $\theta^{3/2}$ by its estimate. The <coverage probability> is asymptotic, not exact at finite $n$. Intersecting with the positive parameter space removes impossible values without changing whether the true positive parameter is covered.
Back to article page
