Confidence regions and minimax rates in outlier-robust estimation on the probability simplex (Q2192314)

The mean of a random vector of the probability simplex is estimated. The minimax rates of the expected error of estimation under the sparsity assumption are established in cases of total variation, Hellinger, Euclidean and Wasserstein distances employed for accuracy measures. The efficincy is illustrated on the dataset which is a collection of books written by Alexandre Dumas and Emile Zola.