The L1-median is a robust estimator of multivariate location with good statistical properties. Several algorithms for computing the L1-median are available. Problem specific algorithms can be used, but also general optimization routines. The aim is to compare different algorithms with respect to their precision and runtime. This is possible because all considered algorithms have been implemented in a standardized manner in the open source environment R. In most situations, the algorithm based on the optimization routine NLM (non-linear minimization) clearly outperforms other approaches. Its low computation time makes applications for large and high-dimensional data feasible.
Algorithm Multivariate median Optimization Robustness
This is a preview of subscription content, log in to check access.
Gower J (1974) As 78: the mediancentre. J R Statist Soc 23: 466–470Google Scholar
Hössjer O, Croux C (1995) Generalizing univariate signed rank statistics for testing and estimating a multivariate location parameter. Nonparametr Statist 4: 293–308MathSciNetzbMATHCrossRefGoogle Scholar
Lopuhaä H, Rousseeuw P (1991) Breakdown points of affine equivariant estimators of multivariate location and covariance matrices. Ann Stat 19(1): 229–248zbMATHCrossRefGoogle Scholar
Nelder J, Mead R (1965) A simplex algorithm for function minimization. Comput J 7: 308–313zbMATHGoogle Scholar