Skip to main content

Timeline for Anti-delta function?

Current License: CC BY-SA 4.0

10 events
when toggle format what by license comment
Feb 1, 2023 at 1:40 comment added Anixx So, using this variant of delta function math.stackexchange.com/questions/4627326/… it would be $\overline{\delta}(1/x)$.
Jan 31, 2022 at 16:37 comment added Willie Wong I think this is basically what Gro-Tsen wrote, except instead of Stone-Čech you take the one-point compactification and require your functions to have a limit at $\infty$.
Jan 30, 2022 at 23:12 history edited LSpice CC BY-SA 4.0
Link to comment
Jan 30, 2022 at 22:41 comment added Anixx As others state that $F(x)$ cannot be a distribution, what $1/x^2 \delta(1/x)$ from your answer supposed to be? A hyperfunction? Or some formal expression?
Jan 30, 2022 at 22:38 comment added Anixx Very interesting, but still I do not understand :-) You said, $F(x)$ is dimensionless. Is it similar to an empty matrix math.stackexchange.com/questions/1913141/… ?
Jan 30, 2022 at 22:32 comment added Carlo Beenakker just dimensional analysis, $F(x)dx$ should be dimensionless, so if you wish to map 0 onto $\infty$ by taking $F(x)\propto\delta(1/x)$ you need a factor $1/x^2$; $\delta_\epsilon(x)$ is a mollified delta function [$\lim_{\epsilon\rightarrow 0}\delta_\epsilon(x)=\delta(x)$].
Jan 30, 2022 at 22:28 comment added Anixx Well, I was wondering what thinking path led you to this development? Also, why it is assumed that the integral $\int_{-\infty}^\infty F_\epsilon(x)=1$ is continuous at $\epsilon=0$? Also, does $\delta_{\epsilon}(x)$ mean delta function with support at $\epsilon$?
Jan 30, 2022 at 22:24 history edited Carlo Beenakker CC BY-SA 4.0
added 2 characters in body
Jan 30, 2022 at 22:23 comment added Anixx Did you use any known properties when deriving this? For instance, identities for $\delta(1/x)$ or Laplace transform?
Jan 30, 2022 at 22:19 history answered Carlo Beenakker CC BY-SA 4.0