[seqfan] Re: linear recurrence question

[Max Alekseyev]

Constrains p*x + q*y + r*z = s, 1 <= x,y,z <= n define a 2-d polygon
in 3-d space.
So a(n) may be even a polynomial in n -- see
http://en.wikipedia.org/wiki/Ehrhart_polynomial
Max

On Sun, Apr 15, 2012 at 2:57 PM, Kimberling, Clark <ck6 at evansville.edu> wrote:
> Hello SeqFans,
>
>
>
> Suppose that p,q,r,s are integers, and let a(n) be the number of triples (x,y,z) such that
>
>
>
> p*x + q*y + r*z = s,
>
>
>
> where x,y,z are all in {1,...,n}.  It appears that (a(n)) is a linear recurrence sequence -- and that x,y,z  can be replaced by x(1),x(2),...,x(k), etc.
>
>
>
> Can someone provide a reference?
>
>
>
> Thanks!
>
>
>
> Clark Kimberling
>
>
> _______________________________________________
>
> Seqfan Mailing list - http://list.seqfan.eu/