Levy distribution in half space based on nonextensive statistical mechanics

Abstract

Probability distributions defined on the half space are known to be quite different from those in the full space. Here, a nonextensive entropic treatment is presented for the half space in an analytic and self-consistent way. In this development, the ordinary first moment of the random variable X is divergent in contrast to the case of the full space. A general (nu)-th moment of X is considered as a constraint in the principle of maximum Tsallis entropy. The infinite divisibility of the distribution with an arbitrary (nu) larger than zero and convergence of its N-fold convolution to the exact Levy-stable distribution is discussed in detail. A feature of this derivation is that the Levy index is related to both the values of (nu) and the index of nonextensivity.