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 #39 Dec 13 2024 16:43:30
%S 1,2,4,5,8,8,13,12,16,17,24,18,27,26,32,31,40,32,45,36,46,51,64,42,57,
%T 58,68,61,78,60,83,68,80,85,100,74,99,94,110,91,116,90,121,104,116,
%U 127,152,100,131,122,144,137,166,130,161,136,162,171,202,126,171,164,182,163,190
%N a(n) = Sum_{k=1..n} phi(floor(n/k)) where phi is the Euler totient function.
%H Robert Israel, <a href="/A317625/b317625.txt">Table of n, a(n) for n = 1..10000</a>
%H Olivier Bordellès, Randell Heyman, and Igor E. Shparlinski, <a href="https://arxiv.org/abs/1808.00188">On a sum involving the Euler function</a>, arXiv:1808.00188 [math.NT], 2018.
%F a(n) <= (1/2)*(1 + 1/zeta(2))*n*log(n) + 4*n + sqrt(n)*log(n)/4 + sqrt(n), uniformly for n >= 3.
%F a(n) >= ((2629/4009)+o(1))*n*log(n)/zeta(2) as n approaches infinity.
%F Cautious conjecture: a(n) ~ n*log(n)/zeta(2).
%e a(4) = phi(floor(4/1))+phi(floor(4/2))+phi(floor(4/3))+phi(floor(4/4)) = phi(4)+phi(2)+phi(1)+phi(1) = 2+1+1+1 = 5.
%p with(numtheory): S:=0: for x to 30 do: for m to x do: S := S+phi(trunc(x/m)) end do; print(x, S); S := 0:end do:
%t Array[Sum[EulerPhi[Floor[#/k]], {k, #}] &, 65] (* _Michael De Vlieger_, Aug 02 2018 *)
%o (PARI) a(n) = sum(x=1, n, eulerphi(n\x)); \\ _Michel Marcus_, Aug 02 2018
%K nonn
%O 1,2
%A _Randell G Heyman_, Aug 02 2018
%E More terms from _Michel Marcus_, Aug 02 2018