2012/09/17 by N. Gagunashvili, N. D. Gagunashvili, M. Schmelling +2
Computer Science · Engineering · Mathematics · Physics and Astronomy · #Distributed Sensor Networks and Detection Algorithms #Fault Detection and Control Systems #astro-ph.IM #hep-ex #msc:62-07 #msc:62F03 #msc:62F10 #msc:62P30 #msc:62P35 #nucl-ex #physics.data-an #stat.AP #stat.CO
paper · pdf · doi:10.48550/arxiv.1209.3766
22 pages, 6 figures
arxiv created 2012/09/17 · arxiv updated 2012/09/19
A kernel based procedure for correcting experimental data for distortions due to the finite resolution and limited detector acceptance is presented. The unfolding problem is known to be an ill-posed problem that can not be solved without some a priori information about solution such as, for example, smoothness or positivity. In the approach presented here the true distribution is estimated by a weighted sum of kernels, with the width of the kernels acting as a regularization parameter responsible for the smoothness of the result. Cross-validation is used to determine an optimal value for this parameter. A numerical example with a simulation study of systematical and statistical errors is presented to illustrate the procedure.