Preliminary results of an interlaboratory study of chemometric software and methods on NIR data. Predicting the content of crude protein and water in forages

摘要: Abstract Within the framework of ‘European Network for Intercomparison Chemometric Software and Methods,’ a project supported by European Commission, we have carried out proficiency study on near infrared spectroscopy (NIR) data set to determine how different results from several laboratories were when they used their preferred multivariate calibration method software. The was distributed six participants, all whom had previous knowledge experience calibration. consisted NIR spectra 305 forage samples recorded under same conditions specialised laboratory. Two parameters predicted, moisture at 103–105 °C crude protein content. Results showed that root mean square error prediction (RMSEP) values obtained acceptable, although varied considerably. These preliminary will be properly define final in which more participants collaborate.