https://www.selleckchem.com/pr....oducts/pemigatinib-i
AIM AND OBJECTIVE Near Infrared (NIR) spectroscopy data are featured by few dozen to many thousands of samples and highly correlated variables. Quantitative analysis of such data usually requires a combination of analytical methods with variable selection or screening methods. Commonly-used variable screening methods fail to recover the true model when (i) some of the variables are highly correlated, and (ii) the sample size is less than the number of relevant variables. In these cases, partial least squares (PLS) regress
Like
Comment
Share
