Time Series Modeling of Adani Power & Tata Power Closed Prices based on Adaptive Neuro-fuzzy Inference System-Wavelet Model

Abstract

This study integrates wavelet transforms with the Adaptive Neuro-Fuzzy Inference System (ANFIS) to forecast the stock prices of Tata Power and Adani Power. Using Daubechies wavelet (Db8) at level 3, the data were decomposed to capture complex patterns. Performance was evaluated using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Squared Error (MSE). In the test phase, the ANFIS- Wavelet model improved RMSE by 5.89%, MAE by 8.35%, and MSE by 11.44% for Tata Power, and RMSE by 6.34%, MAE by 0.38%, and MSE by 12.27% for Adani Power compared to the standard ANFIS model. These results highlight the superior accuracy of the hybrid approach for time series forecasting, offering better insights for decision-making in financial analysis.