Title :
Model-selection-based economic prediction system using MEBML
Author :
Sun, Chengyi ; Sun, Yan ; Sun, Yu
Author_Institution :
Comput. Center, Taiyuan Univ. of Technol., China
fDate :
6/21/1905 12:00:00 AM
Abstract :
This paper suggests a model-selection-based economic prediction system (MSBEPS) using mind-evolutionary-based machine learning (MEBML). MEBML was put forward by the authors (1998) and in MSBEPS it is used to capture time-variant features in economic time series. With MEBML the prediction system is suited to a changing environment. The system can determine the type of model, its order and the coefficients of the basic functions at the same time. In the experiment the solution space is multi-dimensional (9-15 dimensions) and the types of data in each dimension are different. The experiment proves that MEBML is efficient in searching a solution space which consists of different types of data
Keywords :
economic cybernetics; forecasting theory; learning (artificial intelligence); search problems; time series; MEBML; changing environment; economic time series; macroeconomics forecasting; mind-evolutionary-based machine learning; model-selection-based economic prediction system; time-variant features; Econometrics; Economic forecasting; Electronic mail; Environmental economics; Expert systems; Macroeconomics; Neural networks; Predictive models; Sun; Training data;
Conference_Titel :
Systems, Man, and Cybernetics, 1999. IEEE SMC '99 Conference Proceedings. 1999 IEEE International Conference on
Conference_Location :
Tokyo
Print_ISBN :
0-7803-5731-0
DOI :
10.1109/ICSMC.1999.825333
