In the context of the Wald statistic, for a parameter $\theta$ and its maximum likelihood estimator $\hat{\theta}$, what is the correct asymptotic distribution of the statistic $W = \frac{(\hat{\theta} - \theta_0)^2}{Var(\hat{\theta})}$ under the null hypothesis $H_0: \theta = \theta_0$?