Linear regression analysis and its application to the multivariate spectral calibrations for the multiresolution of a ternary mixture of caffeine, paracetamol and metamizol in tablets.

The multivariate spectral calibration methods, tri-linear regression-calibration (TLRC) and multi-linear regression-calibration (MLRC) were developed for the multiresolution of a ternary mixture of caffeine (CAF), paracetamol (APAP), metamizol (MET), which have closely overlapped in the spectra. The calibration algorithms were briefly described for the three-component system, CAF-APAP-MET. By using the various synthetic mixtures of three compounds, the validity of the TLRC and MLRC methods was confirmed and applied to the real samples containing the above-mentioned compounds in two different commercial tablet formulations. The TLCR and MLRC methods which are very rapid, easy to apply, yet not expensive, are powerful tools with very simple mathematical contents for multiresolution of the three- or multi-component mixture systems. The data treatments were carried out by the MAPLE V, EXCEL and SPSS 10.0 Softwares. The obtained results were successfully compared with each other as well as with those obtained by other literature methods.