1
$\begingroup$

In Mathematica:

LaplaceTransform[Exp[-Exp[-t]], t, s]
Out:= Gamma[s]-Gamma[s,1]

But performing InverseLaplaceTransform on the result does not give me any output. Even InverseLaplaceTransform[Gamma[s], s, t] does not give any output, but InverseLaplaceTransform[-Gamma[s,1], s, t] returns Exp[-Exp[-t]] which is identical to my input. Does that mean InverseLaplaceTransform[Gamma[s], s, t]=0 according to Mathematica?

Integration by summing residues shows that inverse Laplace transform of Gamma function is: $$\mathbf{L^{-1}}[\Gamma(s)]=\int_{\gamma-i\infty}^{\gamma+i\infty}ds~e^{st}\Gamma(s)\\ =\sum_{n=0}^\infty Res_{s=-n}[e^{st}\Gamma(s)]\\ =\sum_{n=0}^\infty e^{-nt}\frac{(-1)^n}{n!}=\sum_{n=0}^\infty \frac{(e^{-t})^n}{n!}=e^{-e^{-t}}$$

in which Gamma[s,1] has no role! I have verified by numerically inverting the Laplace transform (using Talbot's method) that presence of Gamma[s,1] indeed doesn't matter at all (i.e. whether I invert Gamma[s] or Gamma[s]-Gamma[s,1] I get the same number and this matches with Exp[-Exp[-t]]). I know this is a broad question, but what is going on with Mathematica here?

| improve this question | |
$\endgroup$
  • $\begingroup$ I confirm that it did not work for me (Mma 11.3, Win7). $\endgroup$ – Alexei Boulbitch Nov 28 '18 at 9:26
1
$\begingroup$

Workaround:

sol = Sum[LaplaceTransform[(-Exp[-t])^j/j!, t, s], {j, 0, Infinity}]

(*Gamma[s, 0, 1] *)

InverseLaplaceTransform[sol, s, t]

(* E^-E^-t *)
| improve this answer | |
$\endgroup$
  • $\begingroup$ My main objection is why InverseLaplaceTransform[-Gamma[s,1], s, t]=Exp[-Exp[-t]] without Gamma[s] which is also a part (and an important part, as integration by residue shows) of its own output? $\endgroup$ – Deep Nov 28 '18 at 11:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.