文摘
Recent advances in nonparametric functional data analysis allow to define the notion of mode for a sample of curves. A kernel-type estimator is proposed for estimating this modal curve. In addition, other centrality curves can be easily extended to the functional case, namely mean and median curves. A nonparametric unsupervised classification method which is also a cluster analysis is investigated. This is based on the comparison between the modal curve with another centrality curve for measuring some heterogeneity index. The main point is to show the good practical behaviour of this hierarchical classification on a sample of radar waveforms registered from the satellite Topex/Poseidon upon the Amazonian basin. In addition, a few theoretical advances are provided.