A054770 (sort of)

[Leonard Smiley]

Christian wrote:

********************
Some sequences begin at term 2 or higher because the previous terms are
all 0 (as in column k of a triangle). Some begin at term 2 because
the value at term 1 is infinity. In rare cases, the sequence begins at
term -1.
********************

... reminding me of this E-mail exchange from last March, 1999:

********************

N. J. A. Sloane wrote:

> i think this is what he was trying to say:
>
> %I A039834
> %S A039834
1,1,0,1,1,2,3,5,8,13,21,34,55,89,144,233,377,610,987,1597,2584,
> %T A039834
4181,6765,10946,17711,28657,46368,75025,121393,196418,317811,514229,
> %U A039834
832040,1346269,2178309,3524578,5702887,9227465,14930352,24157817
> %V A039834
1,1,0,1,-1,2,-3,5,-8,13,-21,34,-55,89,-144,233,-377,610,-987,1597,-2584,

> %W A039834
4181,-6765,10946,-17711,28657,-46368,75025,-121393,196418,-317811,514229,

> %X A039834
-832040,1346269,-2178309,3524578,-5702887,9227465,-14930352,24157817
> %N A039834 a(n+2)=-a(n+1)+a(n) (signed Fibonacci numbers).
> %Y A039834 Cf. A000045.
> %O A039834 0,6
> %K A039834 sign,done
> %A A039834 Alexander Grasser (pyropunk at usa.net)
> %E A039834 Signs corrected by Len Smiley (smiley at math.uaa.alaska.edu)
and
njas.
>
> - the point being that you want to get the Fibs
> themselves if you continue the recurrence further back into the past

Hmm. Then how about %O A039834   -2,6
Just a thought.
Len

****************************

-2,6   -  yes, good!

****************************

... not to be too negative about things!

Cheers,
Len