On completeness and minimality of random exponential system in a weighted Banach space of functions continuous on the real line (Q2431996): Difference between revisions

Let \(\alpha(t)\) be a non-negative convex function on \({\mathbb R}\) satisfying \[ \lim_{| t| \to\infty} t^{-1}\alpha(t)=\infty, \] and let \(C_\alpha\) denote the Banach space of complex continuous functions \(f\) on the real axis with \(f(t)\exp\{-\alpha(t)\}\) vanishing at infinity, in the uniform norm. The authors study completeness and minimality properties in \(C_\alpha\) of random exponential systems \( \{\exp(\lambda_n+i\psi_n)t, n\geq 1\}\), where \(\lambda_n\) are complex numbers and \(\psi_n\) are independent random variables satisfying some restrictions. The main result (Theorem 1.1) may be viewed as a probabilistic version of Malliavin's classical result, see [\textit{P. Malliavin}, Acta Math. 93, 179--255 (1955; Zbl 0067.05104)]. In particular, if the hypothesis of Theorem 1.1 holds, then if the system \( \exp\{\lambda_nt\}\) is complete/incomplete in \(C_\alpha\), then so is \(\{ \exp(\lambda_n+i\psi_n)t\}\) with probability 1.