Isotopic correlation for Pu composition prediction: Multivariate regresssion approach.

Multivariate regression calibration using multiple linear regression (MLR), principle component regression (PCR) and partial least squares regression (PLSR) algorithm was performed on Pu, Pu, Pu and Pu atom% abundances to predict Pu isotopic abundance. The MLR algorithm was found to be the best among these three algorithms. The effect of Pu composition on the Pu abundance prediction was found to be small but significant especially for achieving high accuracy of <0.