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

MaRDI QIDQ6310368zbMATH OpenFDO

Authors Thomas Deschatre

Publication date 28 November 2018

Mathematics Subject Classification ID

Point processes (e.g., Poisson, Cox, Hawkes processes) (60G55) Nonparametric estimation (62G05) Nonparametric regression and quantile regression (62G08) Inference from stochastic processes and fuzziness (62M86) Applications of statistics to actuarial sciences and financial mathematics (62P05)

Abstract: We consider a doubly stochastic Poisson process with stochastic intensity $lambd{a}_t=nqleft(X_{t}ight)$ where $X$ is a continuous It^o semimartingale and $n$ is an integer. Both processes are observed continuously over a fixed period $left[0,Tight]$. An estimation procedure is proposed in a non parametrical setting for the function $q$ on an interval $I$ where $X$ is sufficiently observed using a local polynomial estimator. A method to select the bandwidth in a non asymptotic framework is proposed, leading to an oracle inequality. If $m$ is the degree of the chosen polynomial, the accuracy of our estimator over the H"older class of order $$ is $$ if $$ and it is optimal in the minimax sense if $$. A parametrical test is also proposed to test if $q$ belongs to some parametrical family. Those results are applied to French temperature and electricity spot prices data where we infer the intensity of electricity spot spikes as a function of the temperature.

This page was built for publication: Local polynomial estimation of the intensity of a doubly stochastic Poisson process with bandwidth selection procedure

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6310368)