|
|
A010912
|
|
Pisot sequences E(3,7), P(3,7).
|
|
3
|
|
|
3, 7, 16, 37, 86, 200, 465, 1081, 2513, 5842, 13581, 31572, 73396, 170625, 396655, 922111, 2143648, 4983377, 11584946, 26931732, 62608681, 145547525, 338356945, 786584466, 1828587033, 4250949112, 9882257736, 22973462017, 53406819691, 124155792775
(list;
graph;
refs;
listen;
history;
text;
internal format)
|
|
|
OFFSET
|
0,1
|
|
LINKS
|
|
|
FORMULA
|
a(n) = 3*a(n-1) - 2*a(n-2) + a(n-3) (holds at least up to n = 1000 but is not known to hold in general).
Empirical g.f.: (3-2*x+x^2)/(1-3*x+2*x^2-x^3). - Colin Barker, Feb 19 2012
Since Pisot (1938) showed that E(3,k) always satisfies a linear recurrence, presumably it would not be difficult to prove that the above conjectures are correct. - N. J. A. Sloane, Jul 30 2016
Theorem: a(n) = 3 a(n - 1) - 2 a(n - 2) + a(n - 3) for n>=3. Proved using the PtoRv program of Ekhad-Sloane-Zeilberger, and implies the above conjectures. - N. J. A. Sloane, Sep 09 2016
|
|
MATHEMATICA
|
|
|
PROG
|
(Magma) XY:=[3, 7]; [n le 2 select XY[n] else Ceiling(Self(n-1)^2/Self(n-2)-1/2): n in [1..32]]; // Klaus Brockhaus, Nov 17 2010
(Magma) a:=1; b:=1; c:=1; S:=[]; for n in [1..32] do a+:=b; b+:=c; c+:=a; Append(~S, c); end for; S; // Klaus Brockhaus, Nov 17 2010
(PARI) Vec((3-2*x+x^2)/(1-3*x+2*x^2-x^3) + O(x^30)) \\ Jinyuan Wang, Mar 10 2020
|
|
CROSSREFS
|
See A008776 for definitions of Pisot sequences.
|
|
KEYWORD
|
nonn,easy
|
|
AUTHOR
|
|
|
STATUS
|
approved
|
|
|
|