OFFSET
1,1
COMMENTS
Perfect numbers (A000396) are a subsequence, since they satisfy sigma(m)/m = 2/1 = (sigma(1)+ 1)/1, that is of the form (sigma(d)+1)/d, with sigma being A000203.
Similarly, k-multiperfect numbers satisfy A240923(m) = k-1.
Holdener et al. say that these numbers have a quasi-friendly divisor and prove that such quasi-friendly divisors cannot have more than two distinct prime divisors. - Michel Marcus, Sep 08 2020, clarified by Antti Karttunen, Aug 31 2025
a(68) > 3.2*10^11. - Giovanni Resta, Aug 30 2025
Question: Might it possible to prove that for all n, A001221(a(n)) = 2, e.g., by showing that this is a subsequence of A387406? In any case, that holds for 67 initial terms. - Antti Karttunen, Aug 31 2025
LINKS
Giovanni Resta, Table of n, a(n) for n = 1..67 (first 51 terms from Michel Marcus)
C. A. Holdener and J. A. Holdener, Characterizing Quasi-Friendly Divisors, Journal of Integer Sequences, Vol. 23 (2020), Article 20.8.4.
MAPLE
filter:= proc(n) uses numtheory; local r; r:= sigma(n)/n; numer(r) - sigma(denom(r)) = 1 end proc:
select(filter, [$1..10^5]); # Robert Israel, Aug 07 2014
MATHEMATICA
a240923[n_Integer] :=
Numerator[DivisorSigma[1, n]/n] -
DivisorSigma[1, Denominator[DivisorSigma[1, n]/n]];
a240991[n_Integer] := Flatten[Position[Thread[a240923[Range[n]]], 1]];
a240991[1000000] (* Michael De Vlieger, Aug 06 2014 *)
CROSSREFS
KEYWORD
nonn
AUTHOR
Michel Marcus, Aug 06 2014
STATUS
approved
