On the application of bootstrapping to laser-induced breakdown spectroscopy data

Abstract

The analysis of laser-induced breakdown spectroscopy (LIBS) data heavily relies on statistical approaches. Recently, the application of another well-established statistical tool has started gaining prominence in LIBS analysis: bootstrapping. In the present work, we investigate the impact of bootstrapping on LIBS data collected by various data acquisition strategies and provide some insights into the application of this powerful statistical tool. We show that bootstrapping primarily affects the standard deviation of the recorded spectra, while barely affecting the mean values. This limits the applicability of bootstrapping to specific cases which are briefly discussed.