I'm stuck at this relatively easy looking integral where I have gaussian distributions \begin{equation} \sigma(x,y)=\frac{1}{\sqrt{2\pi\sigma^2}} e^{-\frac{x^2+y^2}{4\sigma^2}} \end{equation} and the integral \begin{equation} I(x)=\int_{-\infty}^{\infty}dx_N \cdots \int_{-\infty}^{\infty}dx_0 \delta(x-\sum_{i=0}^{N}x_i)\sigma(x_0-\mu_0,x_0-\mu'_0)\sigma(x_1-\mu_1,x_1-\mu'_1)\cdots \sigma(x_N-\mu_N,x_N-\mu'_N). \end{equation}

I tried putting this on mathematica and I got a distribution $I(x)$ that is again a gaussian with the variance now scaled with $N$. I did this only for specific $N$ on mathematica and inferred the scaling. Would appreciate if anybody could guide me on how to do this integral for any $N$.

  • $\begingroup$ Have you heard about Central limit theorem? $\endgroup$ – user35952 Oct 4 at 11:02
  • $\begingroup$ @user35952: Not very helpful in this case. It's obvious that the integral is going to result in a gaussian. What is not obvious is the prefactors that emerge after completing the squares and integrating. $\endgroup$ – Juzar Oct 7 at 2:18

Your Answer

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

Browse other questions tagged or ask your own question.