论文部分内容阅读
We present a quasi-optimal sample set for ordinary least squares regression.The quasi-optimal set is designed in such a way that,for a given number of samples,it delivers the near optimal regression model.The quasioptimal set is defined by maximization of a quantity measuring the mutual orthogonality and determinant of the model matrix.We present its theoretical justification,efficient implementation and demonstrate its efficiency via several numerical examples.