0 avis
Non-parametric level set estimation for spatial data
Archive ouverte : Article de revue
Edité par HAL CCSD ; Pushpa Publishing House
International audience. A non-parametric level set estimator of the density of a stationary d-dimensional spatial process is proposed. The estimator is deduced from a non-parametric kernel density estimator. Berry-Esseen bounds are established and used to give consistency results of the kernel level set estimation, derived from that of the kernel density estimate under some mild conditions.