Almost sure local limit theorem for the Dickman distribution (Q722317)

The Dickman function plays a major role in analytic number theory, and more precisely in the theory of integers which possess only a small number of prime factors (so-called smooth numbers). In the present paper, the authors study the Dickman distribution, which is a probability distribution function which arises naturally from the appropriate normalization of the Dickman function. As pointed out by the authors, this distribution appears as a borderline case in the theory of the local and almost sure local limit theorem. \textit{H.-K. Hwang} and \textit{T.-H. Tsai} [Comb. Probab. Comput. 11, No. 4, 353--371 (2002; Zbl 1008.68044)] introduced a probabilistic model for describing the Dickman distribution. In the present paper, the authors prove a correlation inequality for this random model, and use it to deduce local, strong local, and almost sure local limit theorems.