4
$\begingroup$

I want to calculate the infinite sum of Hermite polynomials, which works fine with older versions of Mathematica, but with version 13 it doesn't. The infinite sum is:

Sum[HermiteH[n, x] λ^n/ n! Exp[-(1/2) (x^2) - λ^2 - I ( n + 1/2) t], {n,0,Infinity},Assumptions -> Element[{λ,x,t}, Reals]]

Can someone help? Thank you very much!

$\endgroup$
5
  • 2
    $\begingroup$ To which older version are you referring? $\endgroup$
    – bmf
    Commented Mar 20, 2022 at 6:17
  • $\begingroup$ That series may diverge: DiscreteAsymptotic[ HermiteH[n, x] \[Lambda]^n/ n! Exp[-(1/2) (x^2) - \[Lambda]^2 - I (n + 1/2) t] /. {x -> 0, t -> 1, \[Lambda] -> 1}, {n, Infinity, 1}] results in $$ \frac{2^{n/2} e^{\left(\frac{1}{2}-i\right) n+\left(-1-\frac{i}{2}\right)} n^{-\frac{n}{2}-\frac{1}{2}} \cos \left(\frac{\pi n}{2}\right)}{\sqrt{\pi }}.$$ $\endgroup$
    – user64494
    Commented Mar 20, 2022 at 8:00
  • $\begingroup$ I am not sure, but I think it was Mathematica version 6 in 2007, where the equation gives a solution. $\endgroup$
    – Klaus
    Commented Mar 20, 2022 at 9:06
  • 2
    $\begingroup$ Looks like answer is: E^((-1/2*I)*t - x^2/2 + (2*x*\[Lambda])/E^(I*t) + (-1 - E^((-2*I)*t))*\[Lambda]^2). $\endgroup$ Commented Mar 20, 2022 at 10:49
  • $\begingroup$ Yes, this was the result I got with the sum. $\endgroup$
    – Klaus
    Commented Mar 20, 2022 at 12:11

1 Answer 1

6
$\begingroup$

By factoring the exponential and pulling the factors not dependent on the running index n out of the sum Mathematica still calculates

Exp[-(1/2) (x^2) - \[Lambda]^2 - I (1/2) t] 
Sum[HermiteH[n, x](\[Lambda] Exp[-I t])^n/n!,{n,0,Infinity}]
(* E^(-((I*t)/2)-x^2/2+(2*x*\[Lambda])/E^(I*t)-
\[Lambda]^2 - \[Lambda]^2/E^(2*I*t)) *)
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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