A non-parametric estimation approach in the investigation of spectral statistics

We have used kernel density estimation (KDE) technique to analyze the spectral statistics of nuclear systems with emphasis on the nearest neighbor spacing distribution. The deviations to regular and chaotic dynamics are described by closer distances to Poisson and Wigner limits, respectively which have calculated via Kullback–Leibler divergence measure. The level statistics of nuclei provide empirical evidences for three dynamical symmetry limits of interacting boson model, considering oblate and prolate nuclei. The predictions of KDE technique suggest a considerable reduction in the uncertainties of chaocity degrees and also more regular dynamics in comparison with other estimation methods for considered systems.