The research aims to predict the behavior of ordinary stock prices using the vector basis function model and the hybrid model consisting of (the vector basis function + the fast Fourier transform). The research was based on a purposive sample, which is the Bank of Baghdad from the banking sector, for the period from 1/1/2006 to 12/31/2025, with (235) monthly observations of the bank's closing prices. The data was divided into two parts: 80% for training (188 views) and 20% for testing (47 views). Three main statistical criteria were used for comparison: the mean squared error (MSE), the root mean squared error (RMSE), and the mean absolute error (MAE) The results showed that the hybrid model (RBF+FFT) outperformed the radial basis function (RBF) model alone, achieving the lowest values across all error criteria. The study reached several important conclusions, most notably the superiority of the hybrid model (RBF+FFT) over the RBF model alone. Based on these findings, the study offered several recommendations, the most important being the adoption of the hybrid (RBF+FFT) as a tool to support investment decisions, given its accuracy in predicting price trends compared to traditional .This confirms that incorporating the cyclical components derived from Fourier analysis significantly improves the model's ability to understand and predict actual price behavior with greater accuracy..
Keywords : forecasting