THE LINEAR MINIMAX PREDICTOR IN FINITE POPULATIONS WITH ARBITRARY RANK UNDER QUADRATIC LOSS FUNCTION
Shenghua Yu · Chinese Annals of Mathematics,series A · 2004
This paper revises the usual quadratic loss function suitably. On the basis of this the minimax property of nomogeneous linear prediction functions is studied. The author obtains the unique linear minimax predictor of linear predictable variable in finite populations with arbitrary rank (it is must comprehended the uniqueness in the sense almost everywhere).