A291787 -id:A291787 - cp's OEIS frontend

A291784 a(n) = (psi(n) + phi(n))/2.

1, 2, 3, 4, 5, 7, 7, 8, 9, 11, 11, 14, 13, 15, 16, 16, 17, 21, 19, 22, 22, 23, 23, 28, 25, 27, 27, 30, 29, 40, 31, 32, 34, 35, 36, 42, 37, 39, 40, 44, 41, 54, 43, 46, 48, 47, 47, 56, 49, 55, 52, 54, 53, 63, 56, 60, 58

Offset: 1

N. J. A. Sloane, Sep 02 2017

This is (A001615 + A000010)/2. It is easy to see that this is always an integer.
If n is a power of a prime (including 1 and primes), then a(n) = n, and in any other case a(n) > n. - M. F. Hasler, Sep 09 2017
If n is in A006881, then a(n)=n+1. - Robert Israel, Feb 10 2019

Richard K. Guy, Unsolved Problems in Number Theory, 3rd Edition, Springer, 2004. See Section B41 (page 96 of 2nd ed., pages 147ff of 3rd ed.).

Hugo Pfoertner, Table of n, a(n) for n = 1..10000
Marcin Mazur and Bogdan V. Petrenko, Generalizations of Arnold's version of Euler's theorem for matrices, Japanese Journal of Mathematics, 5:183-189, 2010.
N. J. A. Sloane, Three (No, 8) Lovely Problems from the OEIS, Experimental Mathematics Seminar, Rutgers University, Oct 05 2017, Part I, Part 2, Slides. (Mentions this sequence)

Cf. A000010, A001615, A006881, A291785, A291786, A291787, A291788.

f:= proc(n) local P, p;
  P:= numtheory:-factorset(n);
  n*(mul((p-1)/p, p=P) + mul((p+1)/p, p=P))/2
end proc:
map(f, [$1..100]); # Robert Israel, Feb 10 2019

psi[n_] := If[n == 1, 1, n*Times @@ (1 + 1/FactorInteger[n][[All, 1]])];
a[n_] := (psi[n] + EulerPhi[n])/2;
Array[a, 100] (* Jean-François Alcover, Feb 25 2019 *)

A291784(n)=(eulerphi(n)+n*sumdivmult(n,d,issquarefree(d)/d))\2 \\ M. F. Hasler, Sep 03 2017

Sum_{k=1..n} a(k) = c * n^2 + O(n*log(n)), where c = 21/(4*Pi^2) = 0.531936... . - Amiram Eldar, Dec 05 2023

A291790 Numbers whose trajectory under iteration of the map k -> (sigma(k)+phi(k))/2 consists only of integers and is unbounded.

Original entry on oeis.org

270, 290, 308, 326, 327, 328, 352, 369, 393, 394, 395, 396, 410, 440, 458, 459, 465, 496, 504, 510, 525, 559, 560, 570, 606, 616, 620, 685, 686, 702, 712, 725, 734, 735, 737, 738, 745, 746, 747, 783, 791, 792, 805, 806, 813, 814, 815, 816, 828

Offset: 1

N. J. A. Sloane, Sep 03 2017

nonn

It would be nice to have a proof that these trajectories are integral and unbounded, or, of course, that they eventually reach a fractional value (and die), or reach a prime (which is then a fixed point). (Cf. A291787.) If either of the last two things happen, then that value of n will be removed from the sequence. AT PRESENT ALL TERMS ARE CONJECTURAL.

When this sequence was submitted, there was a hope that it would be possible to prove that these trajectories were indeed integral and unbounded. This has not yet happened, although see the remarks of Andrew R. Booker in A292108. - N. J. A. Sloane, Sep 25 2017

Hugo Pfoertner, Table of n, a(n) for n = 1..82
Sean A. Irvine, Showing how the initial portions of some of these trajectories merge
N. J. A. Sloane, Three (No, 8) Lovely Problems from the OEIS, Experimental Mathematics Seminar, Rutgers University, Oct 05 2017, Part I, Part 2, Slides. (Mentions this sequence)

Cf. A000010, A000203, A289997, A290001, A291789 (the trajectory of 270), A291787, A292108.

For the "seeds" see A292766.

More terms from Hugo Pfoertner, Sep 03 2017

A291789 Trajectory of 270 under repeated application of k -> (phi(k)+sigma(k))/2.

Original entry on oeis.org

270, 396, 606, 712, 851, 852, 1148, 1416, 2032, 2488, 2960, 4110, 5512, 6918, 8076, 10780, 16044, 23784, 33720, 55240, 73230, 97672, 118470, 169840, 247224, 350260, 442848, 728448, 1213440, 2124864, 4080384, 8159616, 13515078, 15767596, 18626016, 29239504, 39012864, 62623600, 92580308

Offset: 0

N. J. A. Sloane, Sep 03 2017

nonn

The ultimate fate of this trajectory is presently unknown. It may reach a fractional value (when it dies), it may reach a prime (which would be a fixed point), it may enter a cycle of length greater than 1, or it may be unbounded. - Hugo Pfoertner and N. J. A. Sloane, Sep 18 2017

Sean A. Irvine, Table of n, a(n) for n = 0..515 [Terms through a(250) from Hugo Pfoertner, terms a(251)-a(356) from N. J. A. Sloane]
Sean A. Irvine, Illustration of A291789 showing a(n+1)/a(n) (red), cumulative mean of a(n+1)/a(n) (green), and power of 2 in a(n) (blue)
Hugo Pfoertner, Illustration of A291789 using a recursive 5th order Butterworth filter with normalized cut-off frequency of 0.1 (0.5<->Nyquist frequency) to smooth the data.
N. J. A. Sloane, Three (No, 8) Lovely Problems from the OEIS, Experimental Mathematics Seminar, Rutgers University, Oct 05 2017, Part I, Part 2, Slides. (Mentions this sequence)
N. J. A. Sloane, "A Handbook of Integer Sequences" Fifty Years Later, arXiv:2301.03149 [math.NT], 2023, p. 14.

Cf. A000010, A000203, A289997, A290001, A291790, A291787, A291804, A291805.

See A291914 and A292108 for the "big picture".

orbit:= proc(n, m) uses numtheory;
  local V,k;
  V:= Vector(m);
  V[1]:= n;
  for k from 2 to m do V[k]:= (sigma(V[k-1])+ phi(V[k-1]))/2 od:
  convert(V,list)
end proc:
orbit(270, 200); # Robert Israel, Sep 07 2017

NestWhileList[If[! IntegerQ@ #, -1/2, (DivisorSigma[1, #] + EulerPhi@ #)/2] &, 270, Nor[! IntegerQ@ #, SameQ@ ##] &, 2, 38] (* Michael De Vlieger, Sep 19 2017 *)

A292108 Iterate the map k -> (sigma(k) + phi(k))/2 starting at n; a(n) is the number of steps to reach either a fixed point or a fraction, or a(n) = -1 if neither of these two events occurs.

Original entry on oeis.org

0, 0, 0, 1, 0, 1, 0, 1, 1, 1, 0, 2, 0, 3, 2, 1, 0, 1, 0, 2, 2, 1, 0, 4, 1, 2, 1, 4, 0, 2, 0, 1, 4, 3, 2, 1, 0, 3, 2, 1, 0, 9, 0, 2, 3, 1, 0, 7, 1, 1, 2, 1, 0, 8, 3, 2, 2, 1, 0, 3, 0, 8, 7, 1, 3, 2, 0, 1, 7, 6, 0, 1, 0, 3, 2, 4

Offset: 1

Hugo Pfoertner and N. J. A. Sloane, Sep 18 2017

nonn

The first unknown value is a(270).
For an alternative version of this sequence, see A291914.
From Andrew R. Booker, Sep 19 2017 and Oct 03 2017: (Start)
Let f(n) = (sigma(n) + phi(n))/2. Then f(n) >= n, so the trajectory of n under f either terminates with a half-integer, reaches a fixed point, or increases monotonically. The fixed points of f are 1 and the prime numbers, and f(n) is fractional iff n>2 is a square or twice a square.
It seems likely that a(n) = -1 for all but o(x) numbers n <= x. See link for details of the argument. (End)

			Let f(k) = (sigma(k) + phi(k))/2. Under the action of f:
14 -> 15 -> 16 -> 39/2, taking 3 steps, so a(14) = 3.
21 -> 22 -> 23, a prime, in 2 steps, so a(21) = 2.

Hugo Pfoertner, Table of n, a(n) for n = 1..269
Andrew R. Booker, Notes on (sigma + phi)/2
Sean A. Irvine, Showing how the initial portions of some of these trajectories merge
N. J. A. Sloane, Three (No, 8) Lovely Problems from the OEIS, Experimental Mathematics Seminar, Rutgers University, Oct 05 2017, Part I, Part 2, Slides. (Mentions this sequence)
N. J. A. Sloane, "A Handbook of Integer Sequences" Fifty Years Later, arXiv:2301.03149 [math.NT], 2023, p. 14.

Cf. A000010, A000203, A289997, A290001, A291790, A291787, A291804, A291805, A291914.

With[{i = 200}, Table[-1 + Length@ NestWhileList[If[! IntegerQ@ #, -1/2, (DivisorSigma[1, #] + EulerPhi@ #)/2] &, n, Nor[! IntegerQ@ #, SameQ@ ##] &, 2, i, -1] /. k_ /; k >= i - 1 -> -1, {n, 76}]] (* Michael De Vlieger, Sep 19 2017 *)

a(n) = 0 if n is 1 or a prime (these are fixed points).

a(n) = 1 if n>2 is a square or twice a square, since these reach a fraction in one step.

A291785 Iterate the map A291784: k -> (psi(k)+phi(k))/2, starting with n, until a power of a prime (A000961) is reached, or -1 if that never happens.

Original entry on oeis.org

1, 2, 3, 4, 5, 7, 7, 8, 9, 11, 11, 16, 13, 16, 16, 16, 17, 23, 19, 23, 23, 23, 23, 47, 25, 27, 27, 47, 29, 47, 31, 32, 83, 83, 83, 83, 37, 47, 47, 47, 41, 83, 43, 47, -1, 47, 47, -1, 49, -1, 83, 83, 53, 83, -1, -1, 59, 59, 59, -1, 61, 83, 83, 64, 83, 83, 67, -1, -1, -1, 71, -1, 73

Offset: 1

N. J. A. Sloane, Sep 02 2017

sign

Primes and prime powers are fixed points under the map f(k) = (psi(k)+phi(k))/2. (If n = p^k, then psi(n) = p^k(1+1/p), phi(n) = p^k(1-1/p), and their average is p^k, so n is a fixed point under the map.)
Since f(n)>n if n is not a prime power, there can be no nontrivial cycles.
Wall (1985) observes that the trajectories of 45 and 50 are unbounded, so a(45) = a(50) = -1.
Also 48 and many more terms seem to have unbounded trajectories. - Hugo Pfoertner, Sep 03 2017.
Obviously any number in the trajectory of a number with unbounded trajectory (in particular that of 45, A291787) again has this property. A291788 is the union of all these. - M. F. Hasler, Sep 03 2017

Richard K. Guy, Unsolved Problems in Number Theory, 3rd Edition, Springer, 2004. See Section B41, p. 147.

C. R. Wall, Unbounded sequences of Euler-Dedekind means, Amer. Math. Monthly, 92 (1985), 587.

Cf. A000010, A001615, A291784, A291786, A291787, A291788.

A291785(n,L=n)={for(i=0,L,isprimepower(n=A291784(n))&&return(n));(-1)^(n>1)} \\ The search limit L=n is only experimental but appears quite conservative w.r.t. known data, cf. A291786. The algorithm assumes that there are no cycles except for the powers of primes. - M. F. Hasler, Sep 03 2017

More terms from Hugo Pfoertner, Sep 03 2017

A291786 a(n) = number of iterations of k -> (psi(k)+phi(k))/2 (A291784) needed to reach a prime or a power of a prime or 1, or -1 if that doesn't happen.

Original entry on oeis.org

0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 3, 0, 2, 1, 0, 0, 3, 0, 2, 2, 1, 0, 6, 0, 1, 0, 5, 0, 4, 0, 0, 9, 8, 7, 6, 0, 5, 4, 3, 0, 5, 0, 2, -1, 1, 0, -1, 0, -1, 6, 5, 0, 4, -1, -1, 2, 1, 0, -1, 0, 4, 3, 0, 3, 2, 0, -1, -1, -1, 0, -1, 0

Offset: 1

N. J. A. Sloane, Sep 02 2017

Primes and prime powers are fixed points under the map f(k) = (psi(k)+phi(k))/2, so in that case we take a(n)=0. (If n = p^k, then psi(n) = p^k(1+1/p), phi(n) = p^k(1-1/p), and their average is p^k, so n is a fixed point under the map.)
Since f(n)>n if n is not a prime power, there can be no nontrivial cycles.
Wall (1985) observes that the trajectories of 45 and 50 are unbounded, so a(45) = a(50) = -1. See A291787, A291788.

Richard K. Guy, Unsolved Problems in Number Theory, 3rd Edition, Springer, 2004. See Section B41, p. 147.

C. R. Wall, Unbounded sequences of Euler-Dedekind means, Amer. Math. Monthly, 92 (1985), 587.

Cf. A000010, A001615, A291784, A291785, A291787, A291788.

A291786(n,L=n)=n>1&&for(i=0,L,isprimepower(n)&&return(i);n=A291784(n));-(n>1) \\ The suggested search limit L=n is only empirical and might require revision. The code also currently assumes that the prime powers are the only cycles. - M. F. Hasler, Sep 03 2017

a(n) = 0 iff n is in A000961. - M. F. Hasler, Sep 03 2017

Initial terms corrected and more terms from M. F. Hasler, Sep 03 2017

A291788 Numbers n whose trajectory under the map k -> (psi(k)+phi(k))/2 (A291784) grows without limit.

Original entry on oeis.org

45, 48, 50, 55, 56, 60, 68, 69, 70, 72, 74, 75, 76, 77, 78, 80, 84, 85, 86, 87, 88, 90, 91, 92, 93, 94, 95, 96, 98, 99, 102, 104, 105, 108, 111, 112, 115, 116, 117, 118, 119, 120, 122, 123, 124, 126, 133, 134, 135, 136, 140, 141, 142, 143, 144, 145, 146, 147, 152

Offset: 1

N. J. A. Sloane, Sep 03 2017, based on data supplied by Hans Havermann

See A291787 (where A291787(m) = 2*A291787(m-7) for m >= 35) for the trajectory of 45.

There is a similar proof that all the terms from 48 though 152 have a trajectory that merges with the trajectory of 45, and so doubles every 7 steps after a certain point. For example, the trajectory of 152 reaches 2^106*33 at step 390, is 2^107*33 at step 397, and thereafter doubles every 7 steps.- N. J. A. Sloane, Sep 24 2017

Cf. A291784, A291785, A291786, A291787.

Terms 104 to 152 added by N. J. A. Sloane, Sep 24 2017

cp's OEIS Frontend

A291784 a(n) = (psi(n) + phi(n))/2.

Views

Author

Keywords

Comments

References

Links

Crossrefs

Programs

Maple

Mathematica

PARI

Formula

A291790 Numbers whose trajectory under iteration of the map k -> (sigma(k)+phi(k))/2 consists only of integers and is unbounded.

Views

Author

Keywords

Comments

Links

Crossrefs

Extensions

A291789 Trajectory of 270 under repeated application of k -> (phi(k)+sigma(k))/2.

Views

Author

Keywords

Comments

Links

Crossrefs

Programs

Maple

Mathematica

A292108 Iterate the map k -> (sigma(k) + phi(k))/2 starting at n; a(n) is the number of steps to reach either a fixed point or a fraction, or a(n) = -1 if neither of these two events occurs.

Views

Author

Keywords

Comments

Examples

Links

Crossrefs

Programs

Mathematica

Formula

A291785 Iterate the map A291784: k -> (psi(k)+phi(k))/2, starting with n, until a power of a prime (A000961) is reached, or -1 if that never happens.

Views

Author

Keywords

Comments

References

Links

Crossrefs

Programs

PARI

Extensions

A291786 a(n) = number of iterations of k -> (psi(k)+phi(k))/2 (A291784) needed to reach a prime or a power of a prime or 1, or -1 if that doesn't happen.

Views

Author

Keywords

Comments

References

Links

Crossrefs

Programs

PARI

Formula

Extensions

A291788 Numbers n whose trajectory under the map k -> (psi(k)+phi(k))/2 (A291784) grows without limit.

Views

Author

Keywords

Comments

Crossrefs

Extensions