Exponential and Gaussian behavior in the tails of multivariate functions (1108.5011v5)

Abstract: We observe that approximate copies of the function $\Lambda {n}:\mathbb{R}^{n}\rightarrow (0,\infty )$ defined by \begin{equation*} \Lambda _{n}(x)=\exp \left( -x{1}-\pi \sum_{i=2}^{n}x_{i}^{2}\right) \end{equation*} appear in the tails of a large class of functions, with properties related to coordinate independence, convexity, homotheticity, and homogeneity. The function $\Lambda {n}$ is an entropy maximizer (on a half-space) that is uniquely determined by a homogeneity condition together with rotational invariance about the $x{1}$ direction and its behavior near the origin. These results are connected to the limiting Poisson point processes found near the edges of large random samples, as well as the conditioning of random vectors on certain rare events, and can be thought of as variations of Laplace's method for estimating integrals.