Year-end appeal: Please make a donation to the OEIS Foundation to support ongoing development and maintenance of the OEIS. We are now in our 61st year, we have over 378,000 sequences, and we’ve reached 11,000 citations (which often say “discovered thanks to the OEIS”).
%I #30 Nov 18 2018 09:50:48
%S 2,2,3,3,3,3,5,5,5,5,7,7,7,7,7,7,7,7,11,11,11,11,13,13,13,13,13,13,13,
%T 13,17,17,17,17,19,19,19,19,19,19,19,19,23,23,23,23,23,23,23,23,23,23,
%U 23,23,29,29,29,29,31,31,31,31,31,31,31,31,31,31,31,31,37,37,37,37,37
%N a(n) = largest prime <= n/2.
%C Also largest prime factor of any composite <= n. E.g., a(15) = 7 since 7 is the largest prime factor of {4,6,8,9,10,12,14,15}, the composites <= 15.
%C Also largest prime dividing A025527(n) = n!/lcm[1,...,n]. [Comment from _Ray Chandler_, Apr 26 2007: Primes > n/2 don't appear as factors of A025527(n) since they appear once in n! and again in the denominator lcm[1,...,n]. Primes <= n/2 appear more times in the numerator than the denominator so they appear in the fraction.]
%C a(n) is the largest prime factor whose exponent in the factorization of n! is greater than 1. - _Michel Marcus_, Nov 11 2018
%H Michael De Vlieger, <a href="/A055377/b055377.txt">Table of n, a(n) for n = 4..10000</a>
%F a(n) = Max(gpf((n+2) mod k): 1 < k < (n+2) and k not prime), with gpf=A006530 (greatest prime factor). - _Reinhard Zumkeller_, Mar 27 2004
%F Where defined, that is for n > 2, a(A000040(n)) = A000040(A079952(n)). - _Peter Munn_, Sep 18 2017
%e n = 10, n! = 3628800, lcm[1,...,10] = 2520, A025527(10) = 1440 = 32*9*5 so a(7) = 5 (offset = 3).
%t Table[Prime@ PrimePi[n/2], {n, 4, 78}] (* _Michael De Vlieger_, Sep 21 2017 *)
%o (PARI) a(n) = precprime(n/2); \\ _Michel Marcus_, Sep 20 2017
%Y Cf. A000040, A000142, A003418, A006530, A007917, A025527, A060265, A079952, A079953.
%K nonn
%O 4,1
%A _Labos Elemer_, Jun 22 2000; _David W. Wilson_, Jun 10 2005
%E More terms from _James A. Sellers_, Jul 04 2000
%E Edited by _N. J. A. Sloane_ at the suggestion of _Andrew S. Plewe_, May 14 2007