1: \begin{abstract} Spectral properties of bounded linear operators play a crucial role in several areas of mathematics and physics, and arguably the most important one is being positive semidefinite. For each self-adjoint, trace-class operator $O$ we construct a monotone increasing sequence $q_n$ which tends to the minimal eigenvalue $\lambda_{\min}$ if $O$ is not positive semidefinite, and to $0$ otherwise. This sequence only depends on the moments of $O$ and a concrete upper estimate of its $1$-norm; we also demonstrate that it can be effectively calculated for a large class of physically relevant operators. As a by-product, we obtain computable estimates for the $1$-norm of $O$, too. 

2: 

3: First assume that $O$ is positive semidefinite. Unfortunately, positivity tests fail to prove this in finitely many steps. However, $q_n$ gives a rigorous, monotone increasing lower estimate for all eigenvalues, providing a quantitative way of measuring positivity. In this case the speed of convergence is $q_n\approx -\frac cn$. 

4: 

5: Now suppose that $O$ is not positive semidefinite. Then $q_n$ monotonically converges to $\lambda_{\min}$ with super-exponential speed. Hence if $q_n$ stabilizes at a negative value, we obtain a strong indication that $O$ is in fact not positive semidefinite. We also construct an easier computable sequence $q_{n,0}$ which fails to be monotone, but converges to $\lambda_{\min}<0$ faster, providing an even better indicator of non-positivity. 

6: \end{abstract}

7: