# Abel transformation

(Redirected from Abel lemma)
A discrete analog of integration by parts, introduced by N.H. Abel in [Ab]. If $a_1, \ldots, a_N$, $b_1, \ldots, b_N$, are given complex numbers and we set $B_n = \sum_{i\leq n} b_i$ then the summation by parts is the identity $\sum_{k=1}^N a_k b_k = a_N B_N - \sum_{k=1}^{N-1} B_k (a_{k+1} - a_k)\, .$ Observe that, if $B_0$ is arbitrarily chosen and we modify the definition of $B_n$ by $B_n = B_0 + \sum_{i\leq n} b_i$ then the identity becomes $\sum_{k=1}^N = a_N B_N - a_1 B_0 - \sum_{k=1}^{N-1} B_k (a_{k+1} - a_k)\, ,$ in analogy with the arbitrarity of an additive constant in the primitive of a function.
If $a_n\to 0$ and $\{B_n\}$ is a bounded sequence, the Abel transformation shows that $\sum_k a_k b_k$ converges if and only if $\sum_k B_k (a_{k+1}-a_k)$ converges, in which case it yields the formula $\sum_{k=1}^\infty a_k b_k = \sum_{k=1}^\infty B_k (a_k - a_{k+1}) - a_1 B_0\, .$ This fact can be used to prove several very useful criteria of convergence of series of numbers and functions (cf. Abel criterion). The Abel transformation of a series often yields a series with an identical sum, but with a better convergence. It is also regularly used to obtain certain estimates (cf. Abel inequality), in particular, for investigations on the rate of convergence of a series.