错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Critical probabilistic characteristics of the Cramér model for primes and arithmetical properties

  • Michel J. G. Weber

摘要

This work is a probabilistic study of the ‘primes’  of the Cramér model, which consists with sums \(S_n =\sum _{i=3}^n \xi _i\) S n = i = 3 n ξ i , \(n\ge 3\) n 3 , where \(\xi _i\) ξ i are independent random variables such that \(\mathbb {P}\{\xi _i= 1\}= 1-\mathbb {P}\{\xi _i= 1\}=1/{\log i}\) P { ξ i = 1 } = 1 - P { ξ i = 1 } = 1 / log i , \(i\ge 3\) i 3 . We prove that there exists a set of integers \(\mathscr {S} \) S of density 1 such that 0.0.1 \(\begin{aligned} \liminf _{ \mathscr {S}\ni n\rightarrow \infty } (\log n)\mathbb {P}\{S_n\ \hbox {prime} \} \ge \frac{1}{\sqrt{2\pi e}\, } , \end{aligned}\) lim inf S n ( log n ) P { S n prime } 1 2 π e , and that for \(b>\frac{1}{2}\) b > 1 2 , the formula 0.0.2 \(\begin{aligned} \mathbb {P}\{S_n\ \text {prime}\, \} \, =\, \frac{ (1+ o( 1) )}{ \sqrt{2\pi B_n } } \int _{m_n-\sqrt{ 2bB_n\log n}}^{m_n+\sqrt{ 2bB_n\log n}} \, e^{- \frac{(t- m_n)^2}{ 2 B_n } }\, \textrm{d}\pi (t), \end{aligned}\) P { S n prime } = ( 1 + o ( 1 ) ) 2 π B n m n - 2 b B n log n m n + 2 b B n log n e - ( t - m n ) 2 2 B n d π ( t ) , in which \(m_n={\mathbb E \,}S_n,B_n=\textrm{Var }\,S_n\) m n = E S n , B n = Var S n , holds true for all \(n\in \mathscr {S}\) n S , \(n\rightarrow \infty \) n . Further we prove that for any \(0<\eta <1\) 0 < η < 1 , and all n large enough and \( \zeta _0\le \zeta \le \exp \big \{ \frac{c\log n}{\log \log n}\big \}\) ζ 0 ζ exp { c log n log log n } , letting \(S'_n= \sum _{j= 8}^n \xi _j\) S n = j = 8 n ξ j , \(\begin{aligned} \mathbb {P}\big \{ S'_n \zeta \hbox {-quasiprime}\big \} \,\ge \, (1-\eta ) \frac{ e^{-\gamma } }{ \log \zeta }, \end{aligned}\) P { S n ζ -quasiprime } ( 1 - η ) e - γ log ζ , according to Pintz’s terminology, where \(c>0\) c > 0 and \(\gamma \) γ is Euler’s constant. We also test which infinite sequences of primes are ultimately avoided by the ‘primes’  of the Cramér model, with probability 1. Moreover we show that the Cramér model has incidences on the Prime Number Theorem, since it predicts that the error term is sensitive to subsequences. We obtain sharp results on the length and the number of occurences of intervals I such as for some \(z>0\) z > 0 , 0.0.3 \(\begin{aligned} \sup _{n\in I} \frac{|S_n-m_n|}{ \sqrt{B_n}}\le z, \end{aligned}\) sup n I | S n - m n | B n z , which are tied with the spectrum of the Sturm-Liouville equation.