The Effect of Statistical and Artificial Intelligence Models in Stock Price Forecasting: A Study on Real-Time Data

Uncertainty, high volatility and complexity in financial markets make price prediction very difficult for investors. The limitations of traditional methods lead investors to more advanced technologies, and in this context, machine learning and deep learning methods stand out as important tools. In this thesis, different forecasting models are created using machine learning and deep learning methods and stock prices are analyzed. During the training and testing phase of the models, a comprehensive analysis was carried out using real stock data. The performance of the models was compared in detail with criteria such as accuracy rates, error metrics and processing times. The results revealed that the LR model is more successful in stock price forecasting. The findings of the study shed light on future studies to make more accurate investment decisions in financial markets.