

A272718


Partial sums of gcdsum sequence A018804.


3



1, 4, 9, 17, 26, 41, 54, 74, 95, 122, 143, 183, 208, 247, 292, 340, 373, 436, 473, 545, 610, 673, 718, 818, 883, 958, 1039, 1143, 1200, 1335, 1396, 1508, 1613, 1712, 1829, 1997, 2070, 2181, 2306, 2486, 2567, 2762, 2847, 3015, 3204, 3339, 3432, 3672, 3805, 4000, 4165
(list;
graph;
refs;
listen;
history;
text;
internal format)



OFFSET

1,2


LINKS

Charles R Greathouse IV, Table of n, a(n) for n = 1..10000
Olivier Bordellès, A note on the average order of the gcdsum function, Journal of Integer Sequences, vol 10 (2007), article 07.3.3.


FORMULA

According to Bordellès (2007), a(n) = (n^2 / (2*zeta(2)))*(log n + gamma  1/2 + log(A^12/(2*Pi))) + O(n^(1+theta+epsilon)) where A ~= 1.282427129 is the GlaisherKinkelin constant A074962, theta is a certain constant defined in terms of the divisor function and known to lie between 1/4 and 131/416, and epsilon is any positive number.
G.f.: (1/(1  x))*Sum_{k>=1} phi(k)*x^k/(1  x^k)^2, where phi(k) is the Euler totient function.  Ilya Gutkovskiy, Jan 02 2017
a(n) = (1/2)*Sum_{k=1..n} phi(k) * floor(n/k) * floor(1+n/k), where phi(k) is the Euler totient function.  Daniel Suteu, May 28 2018


EXAMPLE

The gcdsum function takes values 1,3,5 for n=1,2,3; therefore a(3) = 1+3+5=9.


PROG

(PARI) first(n)=my(v=vector(n), f); v[1]=1; for(i=2, n, f=factor(i); v[i] = v[i1] + prod(j=1, #f~, (f[j, 2]*(f[j, 1]1)/f[j, 1] + 1)*f[j, 1]^f[j, 2])); v \\ Charles R Greathouse IV, May 05 2016


CROSSREFS

Partial sums of A018804.
Sequence in context: A170879 A134578 A189979 * A301255 A009851 A026403
Adjacent sequences: A272715 A272716 A272717 * A272719 A272720 A272721


KEYWORD

nonn,easy


AUTHOR

Gareth McCaughan, May 05 2016


STATUS

approved



