Niemiro, W. Least empirical risk procedures in statistical inference. (English) Zbl 0797.62021 Appl. Math. 22, No. 1, 55-67 (1993). The author considers the asymptotic representation of the minimum of the empirical risk function under the assumption of convexity. The tests of significance for linear hypotheses are derived. Procedures of the type of least absolute deviations (LAD) and least distance are considered. In particular, an application of LAD-related methods to discriminant analysis is discussed. Reviewer: M.Akahira (Ibaraki) Cited in 1 Document MSC: 62F12 Asymptotic properties of parametric estimators 62F05 Asymptotic properties of parametric tests 62H30 Classification and discrimination; cluster analysis (statistical aspects) Keywords:convex minimization; asymptotics; least distances; asymptotic representation; minimum of the empirical risk function; tests of significance for linear hypotheses; least absolute deviations; least distance; discriminant analysis PDFBibTeX XMLCite \textit{W. Niemiro}, Appl. Math. 22, No. 1, 55--67 (1993; Zbl 0797.62021) Full Text: DOI EuDML