Genetic and evolutionary algorithms for time series forecasting

• Author(s): Cortez, Paulo, 1971- ; Rocha, Miguel ; Neves, José
• Date: 2001
• Persistent ID: http://hdl.handle.net/1822/2223
• Origin: RepositóriUM - Universidade do Minho
• Subject(s): Genetic and evolutionary algorithms; Time series forecasting; Time series analysis; ARMA models
• Document Type: Conference Object
• Language: English

Description

Nowadays, the ability to forecast the future, based only on past data, leads to strategic advantages, which may be the key to success in organizations. Time Series Forecasting allows the modeling of complex systems as black-boxes, being a focus of attention in several research arenas such as Operational Research, Statistics or Computer Science. On the other hand, Genetic and Evolutionary Algorithms (GEAs) are a novel technique increasingly used in Optimization and Machine Learning tasks. The present work reports on the forecast of several Time Series, by GEA based approaches, where Feature Analysis, based on statistical measures is used for dimensionality reduction. The handicap of the evolutionary approach is compared with conventional forecasting methods, being competitive.