Thanks for the explanations and advice.
Richard Fateman wrote:
>It appears that nusum is not confused by your effort to confuse it, and
>implicitly solves the problems as though you have 2 j's.
Andrej Vodopivec wrote
nusum(j,j,0,j) = 0+1+2+...+j = j*(j+1)/2.
>Maybe it is confusing that in the last example
> the upper limit for summation is the same as
> summation variable. The example is
> equivalent to nusum(i, i, 0, j).
I was missing the "dummy variables" idea that the sum cannot depend on what
the sum index is called.
Robert Dodier wrote
>You might consider using the function simplify_sum,
>which applies nusum and other methods.
>load(simplify_sum) loads it,
> ? simplify_sum
>shows some information about it.
Thanks for the suggestion.
ted