We will place certain parts of the theory of statistical efficiency into the author’s
operator trigonometry (1967), thereby providing new geometrical understanding
of statistical efficiency. Important earlier results of Bloomfield
and Watson, Durbin and Kendall, Rao and Rao, will be so interpreted. For
example, worse case relative least squares efficiency corresponds to and is
achieved by the maximal turning antieigenvectors of the covariance matrix.
Some little-known historical perspectives will also be exposed. The overall
view will be emphasized.
Gustafson, K. (2005), The geometry of statistical efficiency, Research Letters in the Information and Mathematical Sciences, 8, 105-121