【正文】 Variable x1 x2 x3 x4 x1 x2 x3 x4 group = B, DF = 4 Variable x1 x2 x3 x4 x1 x2 x3 x4 SAS輸出結(jié)果如下： Simple Statistics TotalSample Standard Variable N Sum Mean Variance Deviation x1 10 x2 10 x3 10 x4 10 group = A Standard Variable N Sum Mean Variance Deviation x1 5 x2 5 x3 5 x4 5 group = B Standard Variable N Sum Mean Variance Deviation x1 5 x2 5 x3 5 x4 5 結(jié)果 Pooled Covariance Matrix Information Natural Log of the Covariance Determinant of the Matrix Rank Covariance Matrix 4 The DISCRIM Procedure Pairwise Squared Distances Between Groups 2 _ _ 1 _ _ D (i|j) = (X X )39。 COV (X X ) i j i j Squared Distance to group From group A B A 0 B 0 結(jié)果 F Statistics, NDF=4, DDF=5 for Squared Distance to group From group A B A 0 B 0 Prob Mahalanobis Distance for Squared Distance to group From group A B A B 結(jié)果 H0: U1=U2 The DISCRIM Procedure Pairwise Generalized Squared Distances Between Groups 2 _ _ 1 _ _ D (i|j) = (X j Xj )39。 COV (Xi Xj ) Generalized Squared Distance to group From group A B A 0 B 0 Linear Discriminant Function _ 1 _ 1 _ Constant = Xj39。 COV X j Coefficient Vector = COV Xj Linear Discriminant Function for group Variable A B Constant x1 x2 x3 x4 結(jié)果 Yi（ X） i= 2 The SAS System The DISCRIM Procedure Classification Results for Calibration Data: Resubstitution Results using Linear Discriminan