A formula for the tail probability of a multivariate normal distribution and its applications (Q700147)

Let \(X,X_1,X_2, \dots\) be a sequence of i.i.d. \(d\)-dimensional Gaussian random vectors with mean \(\mu\). The main result of this paper is an exact asymptotic formula for the tail probability of the random variable \(\|X-\mu \|\), where \(\|\cdot \|\) stands for the Euclidean norm. Moreover, this formula is used to establish two asymptotic results for the maximum deviation from the mean \(M_n=\max_{1\leq i\leq n} \|X_i-\mu \|\): (i) the weak convergence to the Gumbel distribution of the maximum \(M_n\) under positive affine transformations, (ii) the precise almost sure rate of growth of \(M_n\) as \(n\to \infty\). The latter result can be used as a diagnostic tool for validating the assumption of multivariate normality.