Topic: Contributions on statistics for functional data Title: Inference for sparse to non-sparse functional data with covariate-adjustments Authors:  Dominik Liebl - University Bonn (Germany) [presenting]
Abstract: The motivation arises from the problem of analyzing the nonlinear price effects of Germany's abrupt nuclear phaseout after the nuclear disaster in Fukushima Daiichi, Japan, in mid-March 2011. The technical side deals with the nonparametric local linear estimation of the mean and covariance function from a latent functional time series of random functions with covariate-adjustments. By contrast to the classical case of sparse functional data, the amount of discretization points per function is allowed to diverge with the number of functions in our asymptotic analysis. This broader asymptotic scenario includes all cases from (very) sparsely to (very) densely sampled discretization points per function and therefore allows to take into account the somewhat intermediate case in our application. The derived bias, variance, and bandwidth results are used to test for differences in the electricity prices before and after Germany's nuclear phaseout for given values of the most important exogenous factors: electricity demand and air temperature.