Abstract:For a Borel set $A$ and a stationary Poisson point process $\eta_t$ in $\mathbb R^d$ of intensity $t>0$, the Poisson-Delaunay approximation $ A_{\eta_t}$ of $A$ is the union of all Delaunay cells generated by $\eta_t$ with center in $A$. It is shown that $\lambda_d(A_{\eta_t})$ is an unbiased estimator for $\lambda_d(A)$, variance bounds and a quantitative central limit theorem are given. The asymptotic behaviour of the symmetric difference $\lambda_d(A\Delta A_{\eta_t})$ is derived as $t \to\infty$.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: how to use the Poisson - Delaunay approximation to estimate the volume of an unknown set A and evaluate the accuracy of this estimation method. Specifically, the authors consider a stationary Poisson point process $\eta_t$ with intensity $t > 0$ in $\mathbb{R}^d$. For a Borel set A, the Poisson - Delaunay approximation $A_{\eta_t}$ is defined as the union of all Delaunay cells whose centers are located within A. The main research objectives include: 1. **Unbiased Estimation**: It is proved that $\lambda_d(A_{\eta_t})$ is an unbiased estimator of $\lambda_d(A)$, that is: \[ E[\lambda_d(A_{\eta_t})]=\lambda_d(A) \] 2. **Variance Bound and Central Limit Theorem**: The upper bound of the variance of $\lambda_d(A_{\eta_t})$ is given, and when $t \to \infty$, its quantitative central limit theorem after standardization is provided, for example: \[ d_K\left(\frac{\lambda_d(A_{\eta_t}) - E[\lambda_d(A_{\eta_t})]}{\sqrt{V[\lambda_d(A_{\eta_t})]}}, N\right) \leq c_{d,A} t^{-1/2 + 1/(2d)} \] where N is a standard Gaussian random variable and $d_K$ represents the Kolmogorov distance. 3. **Asymptotic Behavior of Symmetric Difference**: Analyze the asymptotic behavior of the volume $\lambda_d(A \Delta A_{\eta_t})$ of the symmetric difference between A and $A_{\eta_t}$ when $t \to \infty$, and obtain the decay rate: \[ \lim_{t \to \infty} t^{1/d} E[\lambda_d(A_{\eta_t} \Delta A)] = c_d \text{Per}(A) \] where $\text{Per}(A)$ represents the perimeter of set A (for convex sets, this corresponds to the $(d - 1)$-dimensional Hausdorff measure on the boundary). Through these results, the authors demonstrate the effectiveness and precision of the Poisson - Delaunay approximation in estimating the volume of unknown sets, especially in the case of large samples (i.e., large t). In addition, they also explore the parallelism between this method and the existing Poisson - Voronoi approximation and verify its applicability to a broader class of sets.

Poisson-Delaunay approximation

On the Poisson approximation of random diagonal sums of Bernoulli matrices

Approximation intensity for pairwise interaction Gibbs point processes using determinantal point processes

Couplings and Poisson approximation for stabilizing functionals of determinantal point processes

Poisson intensity parameter estimation for stationary Gibbs point processes of finite interaction range

Poisson approximation for stochastic processes summed over amenable groups

Poisson fluctuations for edge counts in high-dimensional random geometric graphs

Intersections of Poisson k-flats in hyperbolic space: completing the picture

Poisson-Laguerre tessellations

Nonparametric intensity estimation from noisy observations of a Poisson process under unknown error distribution

Asymptotic properties of maximum likelihood estimators for determinantal point processes

Nonparametric estimation of the stationary density for Hawkes-diffusion systems with known and unknown intensity

Stein estimation of the intensity of a spatial homogeneous Poisson point process

Poisson Approximation to Weighted Sums of Locally Dependent Random Variables

From Monte Carlo to neural networks approximations of boundary value problems

Poisson approximation for cycles in the generalised random graph

On a particle approximation to the Dean-Kawasaki type equation with logarithmic interactions

Stochastic Transport with Lévy Noise -- Fully Discrete Numerical Approximation

Approximation of the invariant measure for stable SDE by the Euler-Maruyama scheme with decreasing step-sizes

Approximating partial differential equations without boundary conditions

Strong Approximations for Empirical Processes Indexed by Lipschitz Functions