New simpler bounds to assess the asymptotic normality of the maximum likelihood estimator

The very recent paper Anastasiou and Reinert (2015) has derived explicit bounds for the bounded Wasserstein distance between the exact, unknown distribution of maximum likelihood estimators (MLEs) and the asymptotic normal distribution. In the present paper, we propose a novel approach to this fundamental issue by combining the Delta method, Stein's method, Taylor expansions and conditional expectations, for the situations where the MLE can be expressed as a function of a sum of independent and identically distributed terms. This encompasses in particular the broad exponential family of distributions. We will show that, in all these cases, our bounds improve on (or are at least as good as) the Anastasiou-Reinert bounds in terms of sharpness and simplicity.