The use of neural networks in financial applications has gained enormous popularity in the recent years. By using a data driven empirical analysis, the main goal is to obtain insights into the dynamics of time series and out-of-sample forecasting. Neural networks are widely acknowledged today as an easily “customizable” tool for learning, modeling and studying a lot of problems very difficult to analyze with standard economic models. For instance, they can be used as nonlinear regression models based on a local analysis into clusters, which generalize the standard models used in econometrics and provide an effective tool to capture the main features of price returns, such as fat tails, volatility clustering, persistence, and leverage effects. Some applications focus on the principal processes generating the observed time series and make use of neural networks as nonlinear models that are more suited to identify the behavior of specific prices. On the other hand, rule based neuro-fuzzy systems based on the integration of neural networks and high-level linguistic information, extracted for example by a Web mining process, have been proposed too. The aim of this Special Session is to promote research and reflect the most recent advances of neural networks, including their hybridization with evolutionary computation, fuzzy systems, metaheuristic techniques and other intelligent methods, in a series of practical problems relevant to the interaction between machine learning and financial modeling and forecasting, the main interest being finalized for searching optimal relationships in the area of financial engineering, risk management, portfolio optimization, industrial organization, auctions, searching equilibriums, financial forecasting, market simulation, agent-based computational economics, and many other areas.
Organizzatore e Chairman della Special Session dal titolo “Applications of Neural Networks for Financial Modeling and Forecasting” nella 2014 IEEE International Joint Conference on Neural Networks (IJCNN 2014) / Panella, Massimo. - (2014). (Intervento presentato al convegno IEEE International Joint Conference on Neural Networks (IJCNN 2014) tenutosi a Pechino, Cina nel 6-11 luglio 2014).
Organizzatore e Chairman della Special Session dal titolo “Applications of Neural Networks for Financial Modeling and Forecasting” nella 2014 IEEE International Joint Conference on Neural Networks (IJCNN 2014)
PANELLA, Massimo
2014
Abstract
The use of neural networks in financial applications has gained enormous popularity in the recent years. By using a data driven empirical analysis, the main goal is to obtain insights into the dynamics of time series and out-of-sample forecasting. Neural networks are widely acknowledged today as an easily “customizable” tool for learning, modeling and studying a lot of problems very difficult to analyze with standard economic models. For instance, they can be used as nonlinear regression models based on a local analysis into clusters, which generalize the standard models used in econometrics and provide an effective tool to capture the main features of price returns, such as fat tails, volatility clustering, persistence, and leverage effects. Some applications focus on the principal processes generating the observed time series and make use of neural networks as nonlinear models that are more suited to identify the behavior of specific prices. On the other hand, rule based neuro-fuzzy systems based on the integration of neural networks and high-level linguistic information, extracted for example by a Web mining process, have been proposed too. The aim of this Special Session is to promote research and reflect the most recent advances of neural networks, including their hybridization with evolutionary computation, fuzzy systems, metaheuristic techniques and other intelligent methods, in a series of practical problems relevant to the interaction between machine learning and financial modeling and forecasting, the main interest being finalized for searching optimal relationships in the area of financial engineering, risk management, portfolio optimization, industrial organization, auctions, searching equilibriums, financial forecasting, market simulation, agent-based computational economics, and many other areas.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
