Bias Adjusted Sign Covariance Matrix

The spatial sign covariance matrix (SSCM), also known as the normalized sample (NSCM), has been widely used in signal processing a robust alternative to (SCM). It is well-known that SSCM does not provide consistent estimates of eigenvalues shape (normalized scatter matrix). To alleviate this problem, we propose BASIC (Bias Adjusted SIgn Covariance), which performs an approximate bias correction under assumption samples are generated from zero mean unspecified complex elliptically symmetric distributions (the real-valued case addressed). We then use order develop regularized based estimator, Shrinkage estimator (BASICS), suitable for high dimensional problems, where dimension can be larger than size. assess proposed with several numerical examples well linear discriminant analysis (LDA) classification problem real data sets. simulations show compares competing estimators but advantage being significantly faster compute.