Suojin Wang, Journal of Nonparametric Statistics Editor-in-Chief
The issues in the 2010 volume of the Journal of Nonparametric Statistics (JNPS) cover a wide range of topics, including nonparametric testing for survival data, bootstrap estimation for clustered data, nonparametric functional estimation, time series modeling and prediction, two-sample nonparametric testing, and a review of Spearman’s footrule and Gini’s gamma with complements. There are two special issues—22:5 features papers inspired by the International Seminar on Nonparametric Inference and 22:7 features papers inspired by the Nonparametric Statistics: Refined, Redefined, and Renewed workshop. The editorials for these and all other special issues are freely available.
In 22:7, “Aggregated Wavelet Estimation and Its Application to Ultra-Fast fMRI” by S. Efromovich and Z. A. Valdez-Jasso considers a novel data-driven aggregated wavelet estimator for a nonparametric regression function to achieve better estimation, data-compression, and visualization of functions with different smoothness characteristics at low and high frequencies. Its application to the applied problem of denoising and compression of ultra-fast functional magnetic resonance imaging is particularly interesting.
Another paper, “Testing for Common Principal Components Under Heterokurticity” by M. Hallin, D. Paindaveine, and T. Verdebout is dedicated to the memory of Bernhard Flury. It is shown that Flury’s likelihood ratio test for common principal components can be modified into a pseudo-Gaussian test that remains valid under arbitrary (and possibly heterokurtic) elliptical densities with finite fourth-order moments. Furthermore, this test retains its optimality features in the Gaussian setting.
JNPS is now a publication of the American Statistical Association. ASA members are entitled to a personal online subscription for $15 per volume.
Leave a Reply