I noticed that on Desmos, for $m>0$ and $x\geq0$, $$\int_{-m}^{m}{x\choose t}dt$$ closer and closer approximated $2^x$. So, does $$\int_{-\infty}^{\infty}{x\choose t}dt=2^x$$ Assuming that ${x\choose t}=\frac{\Gamma(x+1)}{\Gamma(t+1)\Gamma(x-t+1)}$.
I know that $\sum_{k\geq0}{x\choose k}=2^x$, but there are, of course, differences between $\int_{-\infty}^{\infty}$ and $\sum_{k\geq0}$.
Can someone prove/disprove this?
I once proved a more general result in my old posting. Here is another, shorter proof.
Before proving this claim, let us show that this leads to the desired conclusion. Indeed, it is easy to check that $x \mapsto \binom{\alpha}{x}$ is square-integrable on $\mathbb{R}$. So by the Fourier inversion formula,
$$ \int_{-\infty}^{\infty} \binom{\alpha}{x} e^{-2\pi i \xi x} \, dx = \begin{cases} \left(1 + e^{-2\pi i \xi}\right)^{\alpha}, & |\xi| < \frac{1}{2} \\ 0, & \text{otherwise} \end{cases} $$
Plugging $\xi = 0$ proves OP's claim.
Proof of Claim. We write $I(\alpha, z)$ for the integral in $\text{(*)}$. Since both sides of $\text{(*)}$ are holomorphic functions in $z$, it suffices to prove that $\text{(*)}$ holds for $z \in (-\infty, 0)$ by the principle of analytic continuation.
Let $z \in (-\infty, 0)$ and write $\gamma$ for the curve $e^{2\pi i \theta}$ with $-\frac{1}{2} \leq \theta \leq \frac{1}{2}$. Then with $\xi = e^{2\pi i \theta}$,
\begin{align*} I(\alpha, z) &= \frac{1}{2\pi i} \int_{\gamma} \frac{\left(1 + \xi \right)^{\alpha}}{\xi^{z+1}} \, d\xi, \end{align*}
Deforming the contour $C$ to wrap the branch cut $(-\infty, 0]$ of the principal logarithm,
\begin{align*} I(\alpha, z) &= \frac{1}{2\pi i} \left( \int_{-1-0^+i}^{-0^+i} \frac{\left(1 + \xi \right)^{\alpha}}{\xi^{z+1}} \, d\xi + \int_{0^+i}^{-1+0^+i} \frac{\left(1 + \xi \right)^{\alpha}}{\xi^{z+1}} \, d\xi \right) \\ &= \frac{e^{i\pi(z+1)} - e^{-i\pi(z+1)}}{2\pi i} \int_{0}^{1} \frac{(1-x)^{\alpha}}{x^{z+1}} \, dx \\ &= \frac{\sin((z+1)\pi)}{\pi} \frac{\Gamma(\alpha+1)\Gamma(-z)}{\Gamma(\alpha+1-z)} \\ &= \binom{\alpha}{z}. \end{align*}
In the last two lines, we utilized the beta function identity and Euler's reflection formula. Therefore $\text{(*)}$ holds for $z \in (-\infty, 0)$ and hence for all $z \in \mathbb{C}$. ////