基于GA算法的组合模型在银行流动性风险预警中的应用研究 Alternative Title Research on Liquidity Risk Early Warning Combination Model Based on GA Algorithm 邓泽军 Thesis Advisor 牛明飞 2015-05-23 Degree Grantor 兰州大学 Place of Conferral 兰州 Degree Name 硕士 Keyword 流动性风险预警 遗传算法（GA） 神经网络组合模型 Abstract 在全球经济一体化、互联网金融大势崛起以及我国金融业对外开放进程的不断向前推进的背景下，我国的商业银行正面临着巨大机遇和挑战。近年来，随着人们对商业银行流动性风险认识加深的同时，对其重视程度也越来越大。因此提前准确了解银行的流动性水平以及预警商业银行的流动性风险等级变得更加重要。本文旨在开发出一套有效准确对商业银行流动性风险进行预警的方法，具有较强的现实意义。本文针对商业银行流动性风险预警问题，在国内外有关流动性风险预警模型研究的基础上，根据非线性组合预测原理，提出通过运用遗传算法（GA）来优化组合模型权系数的方法，构建基于GA算法的商业银行流动性风险预警神经网络组合模型。首先，结合非线性组合预测的思想和遗传算法（GA）的原理，确立了遗传算法优化组合模型权系数的基本思路。然后，选取BP神经网路模型和RBF神经网络模型作为组合模型中的单一模型，构建基于GA算法的商业银行流动性风险预警神经网络组合模型，并且将组合模型预测的绝对误差之和作为适应度函数。最后，通过对BP神经网络模型、RBF神经网络模型以及基于GA算法的组合模型应用结果的比较分析。GA组合模型在预测的逼近效果、模型的稳健性以及流动性风险级别划分准确率上更具优势，可以看出组合模型能有效结合单一模型的优点，能更有效的实现对商业银行流动性风险的预警。 Other Abstract On the background of the global economic integration,the rise of the Internet financial trend as well as the process of our financial industry opening to the outside world moving forward,our country's commercial Banks are facing great opportunities and challenges.In recent years,liquidity risk as one of the main risks faced by commercial Banks,more and more cause the attention of people.So accurately in advance to understand that the bank liquidity level and early warning of commercial Banks liquidity risk level becomes more important.The purpose of this paper is developing a set of effective and accurate method for early warning of commercial Banks liquidity risk that has strong practical significance.As the commercial Banks liquidity risk early warning problem, on the basis of domestic and foreign research on liquidity risk early warning model,according to the principle of non-linear combination forecast and genetic algorithm (GA) to optimize model portfolio weight coefficient method,the construction of commercial bank liquidity risk early warning based on GA algorithm network combination model. First of all,using the thought of the non-linear combination forecast and the principle of genetic algorithm (GA),established the basic ideas of the combination of genetic algorithm to optimize model weights.Then,the BP network model and RBF network model as the combination model of single model, build commercial bank liquidity risk early warning based on GA algorithm network combination model,and the combination model of prediction error is the sum of absolute value as a fitness function.Finally,based on the BP network model,RBF network model, and based on the combination of GA algorithm applied the results of comparative analysis GA combined model in the prediction of approximation effect,the robustness of the model and the level of liquidity risk on the classified accuracy has more advantages,it can be seen that combination model can effectively combine the advantages of single model,can be more effective implementation of commercial Banks liquidity risk early warning. URL 查看原文 Language 中文 Document Type 学位论文 Identifier https://ir.lzu.edu.cn/handle/262010/225231 Collection 数学与统计学院 Recommended CitationGB/T 7714 邓泽军. 基于GA算法的组合模型在银行流动性风险预警中的应用研究[D]. 兰州. 兰州大学,2015.
 Files in This Item: There are no files associated with this item.
 Related Services Recommend this item Bookmark Usage statistics Export to Endnote Altmetrics Score Google Scholar Similar articles in Google Scholar [邓泽军]'s Articles Baidu academic Similar articles in Baidu academic [邓泽军]'s Articles Bing Scholar Similar articles in Bing Scholar [邓泽军]'s Articles Terms of Use No data! Social Bookmark/Share
No comment.