Two multivariate calibration-prediction techniques, principal component regression (PCR) and partial least-squares regression (PLSR) were applied to the chromatographic multicomponent analysis of the drug containing lansoprazole (LAN), clarithromycin (CLA) and amoxicillin (AMO). Optimum chromatographic separation of LAN, CLA and AMO with atorvastatin as the internal standard (IS) was obtained by using Xterra((R)) RP18 column 5 mu m 4.6 x 250 mm(2), and 25 mM ammonium chloride buffer prepared ammonium chloride, acetonitrile and bidistilled water (45:45:10 v/v) as the mobile phase at flow rate 1.0 mL/min. The high pressure liquid chromatography data sets consisting of the ratios of analyte peak areas to the IS peak area were obtained by using diode array detector detection at five wavelengths (205, 210, 215, 220 and 225 nm). LC-chemometric calibration for LAN, CLA and AMO were separately constructed by using the relationship between the peak-area ratio and training sets for each analyte. A series of synthetic solutions containing different concentrations of LAN, CLA and AMO were used to check the prediction ability of the PCR and PLS. Both of the two-chemometric methods in this study can be satisfactorily used for the quantitative analysis and for dissolutions tests of multicomponent commercial drug.