Implements the ARW procedure to compute an adjusted cutoff for squared Mahalanobis distances, then re-estimates location and scatter excluding points beyond the cutoff.
arw_adjustment(x, m0, c0, alpha, pcrit)A list with the following components:
m, the updated location vector after excluding outliers;
c, the updated covariance matrix;
cn, the adjusted cutoff on Mahalanobis distances;
w, a logical vector indicating which observations have distance less than or equal to cn.
A numeric matrix or data frame of observations (rows) by variables (columns), with at least 2 columns.
A numeric vector of initial location estimates (length equal to number of columns in x).
A numeric covariance matrix corresponding to m0.
Numeric; significance level for the chi-square threshold. Defaults to 0.05 if not provided.
Numeric; minimal proportion for the adjusted cutoff. If not provided, it is computed as: \((0.24 - 0.003p)/\sqrt{n}\) if \(p \leq 10\), or \((0.252 - 0.0018p)/\sqrt{n}\) if \(p > 10\).