B0789
Title: Statistical inference for Gaussian processes with small noise asymptotics
Authors: Yasutaka Shimizu - Waseda University (Japan) [presenting]
Abstract: Gaussian processes are considered with unknown mean functions and known covariance kernels. The goal is parametric inference for the mean function when the noise part asymptotically vanishes. A wide class of mean functions is considered under which the likelihood function is written explicitly, and the LAN results are shown under the small noise asymptotics with continuous-time observations. Moreover, the inference under discrete observations is also discussed, under which the asymptotic normality is shown for the quasi-MLE.
