Développement d'un modèle intelligent pour la prédiction de la volatilité financière

Abstract

Assessing the uncertainty associated with asset price fluctuations is crucial in finance, with volatility being a key measure. The ability to predict this volatility is vital for making effective investment decisions, managing risk and pricing products. This research aims to explain the concept of volatility and its importance for financial institutions and global businesses. It provides a comprehensive overview of machine learning techniques, with emphasis on artificial neural networks, in the context of classification problems such as predicting rises and falls in the BIST 100 index. machine learning models and artificial neural networks is detailed, including the process of data collection, data preparation, algorithm selection and model optimization. Each step is essential to develop reliable and efficient models. The models are then implemented and the results analyzed, comparing the performance of different models to identify the best model for predicting market volatility. The research results show that random forest models and decision trees achieve the best results in predicting financial market volatility compared to other models tested. The research results constitute a valuable contribution to the field of forecasting financial market volatility using machine learning and artificial neural network techniques. The study highlights the potential of these techniques to significantly improve forecast accuracy, analyze risks and make sound investment decisions. By demonstrating the effectiveness of machine learning and artificial neural network models in predicting financial market volatility, this study opens vast possibilities for improving investment performance and risk management in finance. The results also suggest the need for further research to develop more complex and accurate models, taking into account the ever-changing economic and financial landscape