Fractional binomial distributions induced by the generalized binomial theorem and their applications

Abstract: A generalized binomial theorem was established by Hara and Hino [Bull. London Math. Soc. 42 (2010), 467-477] for proving the neo-classical inequality. We introduce a new fractional analogue of the binomial distribution on the basis of the generalized binomial theorem and study its properties. They include the law of large numbers, the central limit theorem, and the law of small numbers, the last of which leads to a fractional analogue of the Poisson distribution. As an application of the fractional binomial distribution, a fractional analogue of the Bernstein operator is proposed, which acts on the space of continuous functions on the unit interval. We show that the fractional Bernstein operator uniformly approximates every continuous function on the unit interval. Some limit theorems for its iterates are discussed as well. In particular, a generalization of the Wright--Fisher diffusion process arising in the study of population genetics is captured as a certain limit of the iterates.