摘要：Identification of representative regimes of wave height and direc-tion under di erent wind conditions is complicated by issues that relate tothe specification of the joint distribution of variables that are defined onlinear and circular supports and the occurrence of missing values. We takea latent-class approach and jointly model wave and wind data by a finitemixture of conditionally independent Gamma and von Mises distributions.Maximum-likelihood estimates of parameters are obtained by exploiting asuitable EM algorithm that allows for missing data. The proposed model isvalidated on hourly marine data obtained from a buoy and two tide gaugesin the Adriatic Sea.
关键词：Circular data; cross-validation; EM algorithm; Gamma distri-;bution; latent classes; marine data; missing values; Von Mises distribution.
