Title: Semi- and nonparametric modeling of environmental time series distributions
Authors: Joachim Schnurbus - University of Passau (Germany) [presenting]
Harry Haupt - University of Passau (Germany)
Abstract: Nonparametric regression-based approaches are proposed for modeling the distribution of stochastic processes which may exhibit nonlinearities and non-stationarities driven by trend and/or seasonal patterns. Special emphasis lies on a novel approach for smoothing of potentially multiple seasonal structures. An application to environmental time series reveals that the smoothing approach provides a simple and computationally cheap approach for distribution estimation and forecasting of such processes.