2017/11/09 by Dominik Liebl, Liebl, Dominik · 1 citation
Mathematics · #Statistical Methods and Inference
paper · pdf · doi:10.48550/arxiv.1711.03367
This work is motivated by the problem of testing for differences in the mean\nelectricity prices before and after Germany's abrupt nuclear phaseout after the\nnuclear disaster in Fukushima Daiichi, Japan, in mid-March 2011. Taking into\naccount the nature of the data and the auction design of the electricity\nmarket, we approach this problem using a Local Linear Kernel (LLK) estimator\nfor the nonparametric mean function of sparse covariate-adjusted functional\ndata. We build upon recent theoretical work on the LLK estimator and propose a\ntwo-sample test statistics using a finite sample correction to avoid size\ndistortions. Our nonparametric test results on the price differences point to a\nSimpson's paradox explaining an unexpected result recently reported in the\nliterature.\n