In this study, simultaneous measurements of levodopa (LD), carbidopa (CD), and entacapone (ENT) as anti-parkinson drugs were investigated in synthetic mixtures and Staparkin tablet by continuous wavelet transform (CWT) and radial basis function neural network (RBF-NN) along with spectrophotometric method. In the CWT method, based on the best zero crossing point and R squared (R2) related to the calibration solutions, the Symlet with second order (Sym2), the Daubechies with second order (Db2), and Gaussian as wavelet families were selected for LD, CD, and ENT, respectively. Limit of detection (LOD) and limit of quantification (LOQ) of standard samples for LD, CD, and ENT were 0.0076, 0.159, 0.1519 and 0.0055, 0.0148, 0.3854 μg mL−1, respectively. Also, the average recovery of synthetic mixtures, containing LD, CD, and ENT was obtained 100.23%, 100.01%, and 99.88%, respectively. In the RBF-NN model, root mean square error (RMSE) of 1.14 × 10−27, 2.94 × 10−14, and 8.675 × 10−15 were found for LD, CD, and ENT, respectively. The comparison of proposefd methods with each other was performed on a commercial tablet sample through analysis of variance (ANOVA) test at 95% confidence level. In addition, the standard addition technique was applied to evaluate the performance of these methods in complex matrix. Due to the fast, simple, inexpensive, accurate, as well as without need separation steps, the proposed method can compete with high-performance liquid chromatography (HPLC).