Continuity of an L-series on a closed half-plane of summability #
If a Dirichlet series is summable at s, then at every point z with s.re ≤ z.re its terms
have norms at most those of the terms at s, so the series converges uniformly on the closed
half-plane {z | s.re ≤ z.re} and LSeries a is continuous there. On the vertical line
s + ℝ * I through s the norms even agree exactly, and continuity along that line is a special
case of the half-plane statement.
Mathlib's LSeries_differentiableOn gives more, but only strictly inside the half-plane of
absolute convergence: it needs abscissaOfAbsConv a < s.re, whereas LSeriesSummable a s only
gives abscissaOfAbsConv a ≤ s.re. The line through a point of summability may therefore be the
boundary line of that half-plane, which is exactly the situation in the Wiener--Ikehara argument.
Main results #
TauCeti.LSeries.continuousOn_LSeries:LSeries ais continuous on the closed half-plane{z | s.re ≤ z.re}wheneverLSeriesSummable a s.TauCeti.LSeries.continuous_LSeries_vertical:fun t : ℝ ↦ LSeries a (s + t * I)is continuous wheneverLSeriesSummable a s.TauCeti.LSeries.tendsto_LSeries_nhdsGT: the real one-sided limit ofLSeries aat a real point of summability is the value there.
A Dirichlet series summable at s converges uniformly on the closed half-plane
{z | s.re ≤ z.re}, hence is continuous there.
Mathlib's LSeries_differentiableOn gives more on the open half-plane cut out by the abscissa
of absolute convergence, but says nothing on its boundary line, which is where the
Wiener--Ikehara argument works.
A Dirichlet series summable at s is continuous along the vertical line through s.
Approaching a real point of summability from the right along the real axis, the values of a Dirichlet series converge to its value there.