Abstract: In view of the weaknesses of BP neural network for economic forecasting，a new and more effective economic forecasting model called Immune Artificial Fish Swarm Algorithm Neural Network（IAFSANN） is developed.IAFSANN can improve the precision of learning，convergence rate and generalization ability through training neural networks，furthermore can overcome the shortcomings of BP neural network to some degree.According to the economic data of Zhanjiang，Guangdong，neural networks have been trained by adopting IAFSA to build an IAFSA-NN，which is realized by MATLAB 7.0 and employed to forecast GDP.The forecasting results are satisfactory，proving that IAFSANN is superior to BP neural network.Meanwhile，IAFSANN turns out to be valid and feasible for economic forecasting.

Key words: immune artificial fish swarm algorithm, Back Propagation（BP） neural network, economic forecasting

摘要： 针对BP神经网络在经济预测存在的问题，提出了一种新的经济预测模型──免疫人工鱼群神经网络（IAFSA-NN）。通过免疫人工鱼群算法（IAFSA）训练神经网络，能显著提高网络的学习精度、收敛速度、泛化能力、还能在一定程度上克服BP神经网络的缺陷。以广东省湛江市的经济数据进行建模，给出了IAFSA训练神经网络的基本原理和步骤，构建了一个免疫人工鱼群神经网络的GDP预测模型，并运用MATLAB7.0进行仿真。实证表明，该模型预测结果优于BP网络预测方法，更接近实际数据，IAFSA神经网络用于经济预测是有效可行的。

关键词: 免疫人工鱼群算法, BP神经网络, 经济预测