Standard deviations and octogonal numbers

[creigh at o2online.de]

Many thanks to Wolfdieter Lang- who submitted A098301 only 
a few days ago....

It seems that:

4*A007655(n+1) + A046184(n) = A055793(n+2) + A098301(n+1) 

A007655, Standard deviation of A007654.
A007654. Numbers n such that standard deviation of 1,...,n is an integer.
A046184, Indices of octagonal numbers which are also square.
A055793, Numbers n such that n and floor[n/3] are both squares; i.e. squares 
which remain squares when written in base 3 and last digit is removed.
A098301, Member r=16 of the family of Chebyshev sequences S_r(n) defined 
in A092184.

(this is only a conjecture)

Sincerly, 
Creighton 

p.s. A sequence from a "symmetry"  thought up last night
(don't know what to do with these just yet...) and subsequently tested 
on the first floretion I could think up: 
2jesseqsig: -1, 0, 1, 1, 3, 7, 7, 7, 15, 15, -1, -1, -1, -65, -129, -129, 
-257, -513, -513, -513, -1025, -1025, -1, -1, -1, 4095, 8191, 8191, 16383, 
32767, 32767 (in which case it seems all numbers are just one number away 
from a power of two).