Subset-Selection Procedures Based on Linear Rank-Order Statistics
H. Büringer, H. Martin, K.-H. Schriever · Birkhäuser Boston eBooks · 1980
Let R=(R 1 ,R 2 ,...,R m ) be a random vector as described in the introduction, associated with the random vector (X 1 ,X 2 ,...,X m ). The general linear rank-order-statistic based on R is defined by where the constants c i , iε{1,...,m}, are called regression constants. The function a:{1,2....,m} → IR yields the so-called scores a(i).We sometimes write a i instead of a(i). These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.