About the asymptotic normality of the maximum likelihood estimator of the binomial distribution

94 Views Asked by At

Question

r.v. $X \sim B(n, \theta)$.

I want to show that $\displaystyle\frac{\sqrt{n}(\hat{\theta}-\theta)}{\sqrt{\theta(1-\theta)}}$ converges in distribution to the standard normal distribution when $n\to \infty$.

What I know

From the asymptotic normality of the maximum likelihood estimator, the following holds

\begin{align} \sqrt{n}(\hat\theta - \theta) \overset{L}{\to} N(0, 1/\mathbb{I}_\theta) \end{align}

where $\mathbb{I}_\theta$ is the Fisher information.

\begin{align} \mathbb{I}_\theta = \frac{n}{\theta(1-\theta)}. \end{align} \begin{align} \therefore \sqrt{n}(\hat\theta - \theta)\overset{L}{\to} N(0, \frac{\theta(1-\theta)}{n})\\ \therefore \frac{\sqrt{n}(\hat\theta - \theta)}{\sqrt{\frac{\theta(1-\theta)}{n}}} \overset{L}{\to} N(0,1)\ ??? \end{align}