# A002449 as Solution to System of Equations?

Paul D. Hanna pauldhanna at juno.com
Thu Mar 15 08:17:25 CET 2007

```Seqfans,
Help is needed to simplify the following problem.

Consider the infinite system of simultaneous equations:
A = 1 + xAB
B = 1 + xBCD
C = 1 + xCDEF
D = 1 + xDEFGH
E = 1 + xEFGHIJ
F = 1 + xFGHIJKL
...
What is the unique solution to the variables A,B,C,... as functions in x?

Is there a way to further simplify the above system of equations
to minimize the computational effort to calculate the
sequences for A,B,C,...?   Ex., B = (A-1)/(xA), etc.

I computed the initial coefficients of A to be:
[1,1,2,6,26,166,1626,25510,664666,29559718,...]
but time did not permit me to get more terms since
the above system of equations (in this raw form)
requires (n+1) iterations of the first 2^n equations
involving 2^(n+1) power series A,B,C,... O(x^(2^n))
just for accuracy on the coefficient of x^n in A!

I believe that A = A(x) is the g.f. of A002449
(number of different types of binary trees of height n):
[1,1,2,6,26,166,1626,25510,664666,29559718,2290267226,...]
due to the coincidence of the initial terms, as well as
the binary nature of the system of equations.

Could someone calculate more coefficients of A
to see if they continue to agree with A002449?

Thanks,
Paul
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.seqfan.eu/pipermail/seqfan/attachments/20070315/77b40cbd/attachment.htm>
```