Different chemometric methods such as classical least squares (CLS), principal components regression (PCR) and partial least squares with one dependent variable (PLS-1) applied on UV spectral data (0 D) and on their first derivatives (1 D) were evaluated for the simultaneous quantification of samples containing mixtures of amiloride hydrochloride, atenolol, hydrochlorothiazide and timolol maleate. Their performances were compared by means of ANOVA tests, which evidenced that 0 D-PCR, 0D-PLS-1, 1D-PCR, 1D-PLS-1, were reproducible and gave statistically similar results, while 0 D-CLS and 1D-CLS displayed higher variances than the former and failed to comply with the Levene's variance homogeneity test at different stages of the method comparison and validation process. The four statistically equivalent procedures were successfully applied to the analysis of synthetic samples with two to four analytes and to commercial tablet preparations containing amiloride hydrochloride and hydrochlorothiazide alone or in association with atenolol or timolol maleate.