Let $m,n$ be two positive integers with $0 < m < n$.
Can we integrate this:
$$I = \int_{-\infty}^\infty \mathrm{d}x_1 \dots \int_{-\infty}^\infty \mathrm{d}x_n \left(\sum_{i=1}^n x_i^2\right)^{-m/2} \exp\left(-\sum_{i=1}^n x_i^2\right)$$
If a closed analytical expression is not possible, what I really need is to evaluate a large deviation limit of the form:
$$\lim_{n\rightarrow\infty} \frac{1}{n} \log I$$
where it is assumed that the ratio $m/n$ remains fixed.
Note that $I$ resembles a negative moment of the radious of a multivariate Gaussian.
I asked a very similar question here: Integrate: $\int_0^1 ||\vec x||^{-m}\mathrm{d}\vec{x}$. But note that this integral has different limits and a decaying quadratic exponential. Maybe I have better luck with this variant ;)
Consider the homomorphism $x \mapsto \left( \|x\|, \dfrac{x}{\|x\|} \right)$ between $\mathbf{R}^d \setminus \{0\}$ and $(0, \infty) \times \mathbf{S}_{d-1}.$ By the change of measures formula, the integral becomes $$\begin{align*} \int\limits_0^\infty dr \int\limits_{\mathbf{S}_{d-1}} d\sigma_d\ e^{-r^2}r^{-m} r^{d-1} &= \sigma_{d-1}(\mathbf{S}_{d-1}) \int\limits_{0}^\infty dr\ e^{-r^2} r^{d-m-1} \\ &= \dfrac{\sigma_{d-1}(\mathbf{S}_{d-1})}{2} \int\limits_0^\infty dr\ e^{-r} r^{\frac{d-m-2}{2}} \\ &= \dfrac{\sigma_{d-1}(\mathbf{S}_{d-1})}{2} \Gamma\left(\dfrac{d-m}{2}\right). \square \end{align*}$$