Multilayer feedforward networks are universal approximators
Neural Networks
The nature of statistical learning theory
The nature of statistical learning theory
A process model to develop an internal rating system: sovereign credit ratings
Decision Support Systems
Modeling on Oil-Importing Risk under Risk Correlation
CSO '09 Proceedings of the 2009 International Joint Conference on Computational Sciences and Optimization - Volume 02
CSO '11 Proceedings of the 2011 Fourth International Joint Conference on Computational Sciences and Optimization
Hi-index | 0.00 |
From the perspective of energy security, this paper focuses on country risk forecasting for major oil exporting countries. Due to the two main characteristics of country risk of oil exporting countries, i.e. the complexity and the mutability, this study proposes a decomposition hybrid approach (DHA) for predicting country risk of oil exporters, based on the principle of ''decomposition and ensemble'' and the strategy of ''divide and conquer''. In DHA, effective decomposition methods, such as ensemble empirical mode decomposition (EEMD), are specially introduced to decompose oil exporter's country risk into a series of relatively easily forecasting components; powerful prediction tools, such as least squares support vector regression (LSSVR), are then implemented to predict all extracted components; and finally these predicted results are fused into an ensemble for the original data via ensemble approaches, such as LSSVR model or simple addition (ADD) approach. Experimental results, with ten major oil exporters as study samples, demonstrate that DHA with decomposition process can be statistically proved to be much stronger and more robust than other popular prediction models.