English

Probabilistic Proofs of Some Generalized Mertens' Formulas Via Generalized Dickman Distributions

Number Theory 2018-10-09 v3 Probability

Abstract

The classical Mertens' formula states that pN(11p)1eγlogN, \prod_{p\le N}\big(1-\frac1p)^{-1}\sim e^\gamma\log N, where the product is over all primes pp less than or equal to NN, and γ\gamma is the Euler-Mascheroni constant. By the Euler product formula, this is equivalent to either of the following statements: i.limNn:pnpN1nnN1n=eγ  ii.n:pnpN1neγlogN. \begin{aligned} &i. \lim_{N\to\infty}\frac{\sum_{n:p|n\Rightarrow p\le N}\thinspace\frac1n}{\sum_{n\le N}\frac1n}=e^\gamma\ \ &ii. \sum_{n:p|n\Rightarrow p\le N}\thinspace\frac1n\sim e^\gamma\log N. \end{aligned} Via some random integer constructions and a criterion for weak convergence of distributions to so-called generalized Dickman distributions, we obtain some generalized Mertens' formulas, some of which are new and some of which have been proved using number-theoretic tools. For example, in the spirit of (i), we show that if AA is a subset of the primes which has natural density θ(0,1]\theta\in(0,1] with respect to the set of all primes, then limNn:pnpNandpA1nnN:pnpA1n=eγθΓ(θ+1), \lim_{N\to\infty}\frac{\sum_{n:p|n\Rightarrow p\le N\thinspace\text{and}\thinspace p\in A}\frac1n} {\sum_{n\le N:p|n\Rightarrow p\in A}\frac1n}=e^{\gamma\theta}\Gamma(\theta+1), and also, for any k2k\ge2, limNn:pnpNandpA(k)1nnN:pnpA(k)1n=eγθΓ(θ+1), \lim_{N\to\infty}\frac{\sum^{'(k)}_{n:p|n\Rightarrow p\le N\thinspace\text{and}\thinspace p\in A}\frac1n} {\sum^{'(k)}_{n\le N:p|n\Rightarrow p\in A}\frac1n}=e^{\gamma\theta}\Gamma(\theta+1), where (k)\sum^{'(k)} denotes that the summation is restricted to kk-free positive integers. In the spirit of (ii), we show for example that n:pnpN(k)1n{(k1)free}ϕ(n{(k1)power})eγlogN, \sum^{'(k)}_{n:p|n\Rightarrow p\le N}\frac1{n_{\{(k-1)-\text{free}\}}\phi(n_{\{(k-1)-\text{power}\}})}\sim e^\gamma\log N, where ϕ\phi is the Euler totient function, and n{(k1)free}n_{\{(k-1)-\text{free}\}} and n{(k1)power}n_{\{(k-1)-\text{power}\}} are the (k1)(k-1)-free part and the (k1)(k-1)-power part of nn.

Keywords

Cite

@article{arxiv.1809.04888,
  title  = {Probabilistic Proofs of Some Generalized Mertens' Formulas Via Generalized Dickman Distributions},
  author = {Ross G. Pinsky},
  journal= {arXiv preprint arXiv:1809.04888},
  year   = {2018}
}

Comments

A few editorial changes were made