An MC2 linear programming approach to combined forecasting
Mathematical and Computer Modelling: An International Journal
A multi-dimensional forward selection method for firms' credit sale
Computers & Mathematics with Applications
A multi-objective decision-making method for commercial banks loan portfolio
CASDMKM'04 Proceedings of the 2004 Chinese academy of sciences conference on Data Mining and Knowledge Management
A multi-factors evaluation method on credit evaluation of commerce banks
CASDMKM'04 Proceedings of the 2004 Chinese academy of sciences conference on Data Mining and Knowledge Management
Hi-index | 0.98 |
In this paper, we describe a mathematical framework of estimating the error bounds of optimal combined forecasting (OCF) and the super-bound of nonoptimal combined forecasting errors. We also determine the existent interval [@l"m"i"n/n, @l"m"a"x/n] of the minimal errors square sum J^* of optimal combined forecasting. Furthermore, we give the relationships between the errors of the general combined forecasting method (including the simple average method) and the errors of each forecasting model in the combination by mathematics analysis.