The Bernstein and Skitovič-Darmois characterization theorems for Gaussian distributions on groups, symmetric spaces, and quantum groups (Q1379595): Difference between revisions

By the well-known Bernstein theorem, if \(\xi\) and \(\eta\) are independent random variables, then \(\xi+\eta\) and \(\xi-\eta\) are independent iff \(\xi\) and \(\eta\) are Gaussian. This theorem was generalized by Skitovich and Darmois. They proved that independence of linear forms \(L_1= \alpha_1\xi_1+\cdots+ \alpha_n\xi_n\) and \(L_2= \beta_1\xi_1+\cdots+ \beta_n\xi_n\), \(n\geq 2\), where \(\xi_j\) are independent random variables and \(\alpha_j\), \(\beta_j\neq 0\), implies that each \(\xi_j\) has a Gaussian distribution. The authors give a survey of possible generalizations of these theorems on groups (Abelian and non-Abelian), symmetric spaces, and quantum groups.