Hello, Does any one know an R-function that performs kernel density estimation when there are many variables ? My data has up to 50 variables and several thousand data points. Does any one know an alternative method to perform nonparametric density estimation on high dimensional data ? Many thanks for the help Serge Zaugg --