Abstract:We investigate the nonparametric estimation problem of the density $\pi$, representing the stationary distribution of a two-dimensional system $\left(Z_t\right)_{t \in[0, T]}=\left(X_t, \lambda_t\right)_{t \in[0, T]}$. In this system, $X$ is a Hawkes-diffusion process, and $\lambda$ denotes the stochastic intensity of the Hawkes process driving the jumps of $X$. Based on the continuous observation of a path of $(X_t)$ over $[0, T]$, and initially assuming that $\lambda$ is known, we establish the convergence rate of a kernel estimator $\widehat\pi\left(x^*, y^*\right)$ of $\pi\left(x^*,y^*\right)$ as $T \rightarrow \infty$. Interestingly, this rate depends on the value of $y^*$ influenced by the baseline parameter of the Hawkes intensity process. From the rate of convergence of $\widehat\pi\left(x^*,y^*\right)$, we derive the rate of convergence for an estimator of the invariant density $\lambda$. Subsequently, we extend the study to the case where $\lambda$ is unknown, plugging an estimator of $\lambda$ in the kernel estimator and deducing new rates of convergence for the obtained estimator. The proofs establishing these convergence rates rely on probabilistic results that may hold independent interest. We introduce a Girsanov change of measure to transform the Hawkes process with intensity $\lambda$ into a Poisson process with constant intensity. To achieve this, we extend a bound for the exponential moments for the Hawkes process, originally established in the stationary case, to the non-stationary case. Lastly, we conduct a numerical study to illustrate the obtained rates of convergence of our estimators.

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

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

On High dimensional Poisson models with measurement error: hypothesis testing for nonlinear nonconvex optimization

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

Asymptotic Properties of Nonparametric Intensity Estimation for Replicated Spatial Point Patterns

Minimum contrast for the first-order intensity estimation of spatial and spatio-temporal point processes

Local polynomial estimation of the intensity of a doubly stochastic Poisson process with bandwidth selection procedure

Bayes inference for a non-homogeneous Poisson process with power intensity law (reliability)

Modelling for Poisson process intensities over irregular spatial domains

Nonparametric density estimation for stationary processes under multiplicative measurement errors

Estimation for the Prediction of Point Processes with Many Covariates

Semi-Parametric Inference for Doubly Stochastic Spatial Point Processes: An Approximate Penalized Poisson Likelihood Approach

Nonparametric estimation of the mixing density using polynomials

Adaptive estimating function inference for non-stationary determinantal point processes

Bayesian nonparametric nonhomogeneous Poisson process with applications to USGS earthquake data

An Adaptive Importance Sampling for Locally Stable Point Processes

Infill Asymptotics for A Stochastic Process Model with Measurement Error

Erlang mixture modeling for Poisson process intensities

Nonparametric estimation of mark's distribution of an exponential Shot-noise process