English

On the arithmetic average of the first $n$ primes

Number Theory 2025-07-17 v3

Abstract

The arithmetic average of the first nn primes, pˉn=1ni=1npi\bar p_n = {1\over n} \sum_{i=1}^n p_i, exhibits very many interesting and subtle properties. Since the transformation from pnpˉnp_n \to \bar p_n is extremely easy to invert, pn=npˉn(n1)pˉn1p_n = n\bar p_n - (n-1)\bar p_{n-1}, it is clear that these two sequences pnpˉnp_n \longleftrightarrow \bar p_n must ultimately carry exactly the same information. But the averaged sequence pˉn\bar p_n, while very closely correlated with the primes, (pˉn12pn\bar p_n \sim {1\over2} p_n), is much "smoother'', and much better behaved. Using extensions of various standard results I shall demonstrate that the prime-averaged sequence pˉn\bar p_n satisfies prime-averaged analogues of the Cramer, Andrica, Legendre, Oppermann, Brocard, Fourges, Firoozbakht, Nicholson, and Farhadian conjectures. (So these prime-averaged analogues are not conjectures, they are theorems.) The crucial key to enabling this pleasant behaviour is the "smoothing'' process inherent in averaging. Whereas the asymptotic behaviour of the two sequences is very closely correlated the local fluctuations are quite different.

Keywords

Cite

@article{arxiv.2505.04951,
  title  = {On the arithmetic average of the first $n$ primes},
  author = {Matt Visser},
  journal= {arXiv preprint arXiv:2505.04951},
  year   = {2025}
}

Comments

V1: 10 pages. V2: now 17 pages. Considerable additional discussion, several bounds strengthened and simplified, minor typos fixed, one reference added. Note particularly the addition of analyses of the Oppermann, Brocard, Fourges, Nicholson, and Farhadian conjectures. V3: Still 17 pages. Minor typos fixed. Closely resembles published version