【作者】 傅强；

【Author】 Fu Qiang(Shanghai University of Economics & Finance, ’98 BS graduate,Shanghai 200041)

【机构】 上海财经大学；

【摘要】 本文运用了人工神经网络的反向传播算法,来实现对国内生产总值的快速估算。文中详细介绍了BPN网络结构的设计、网络训练的具体设计以及实际的估算结果,并对网络能力和性能作出了初步评价。更多还原

【Abstract】 Backpropagation is used as a systematic method for training multilayer artificial neural networks in this paper to build GDP prediction models. Directed search algorithms, such as selecting the number of nodes via sequential network construction (SNC), sensitivity based pruning (SBP) of inputs and nonlinear cross validation (NCV) are employed to search the space of possible architectures based on estimates of generalization performance.更多还原