# Bootstrapping from Distribution Element Trees Smooth bootstrapping or generation of (un)conditional random vectors based on an existing distribution element tree (DET). ```r det.rnd(n, det, xc = vector("numeric", length = 0), dc = vector("numeric", length = 0), cores = Inf) ``` ## Arguments - `n`: number of samples to generate. - `det`: distribution element tree object resulting from `det.construct`. - `xc`: vector with conditioning values of probability-space components listed in `dc`. If empty (default), unconditional samples are generated. - `dc`: integer vector with indices of conditioning components corresponding to `xc`. If empty (default), unconditional samples are generated. - `cores`: for large `n`, `cores > 1` allows for parallel bootstrapping using the indicated number of cores. The default is `cores = Inf`, which allocates half of the available cores (see `detectCores`). `cores = 1` corresponds to serial bootstrapping. ## Returns A matrix containing `n` random vectors (columns) with `d` components or dimensions (rows) is returned. `d` is equal to the dimensionality of the underlying `det` object. ## Examples ```r ## 2d example require(stats); require(graphics) # data from uniform distribution on a wedge x <- matrix(runif(2e4), ncol = 2); x <- x[x[,2]

det.rnd function