基于粗集—神经网络的宏观经济预警研究
发布时间：2018-10-16 22:55
【摘要】： 本文首先对国内外宏观经济预警进行概述，，提出目前在宏观经济预警方面存在的主要问题；并在此基础上阐述宏观经济预警理论体系，讨论了预警指标的设计、预警方法的选择和预警警限的界定等问题；在分析和讨论人工神经网络(ANN)预警方法和粗集(RS)理论方法的优势互补后，建立了基于粗集与神经网络相结合宏观经济预警系统模型，详细地论述了基于粗集-神经网络预警方法的基本原理，研究了预警方法的实现，其中包括建立粗集-神经网络预警的结构模式、解决粗集-神经网络预警操作中的问题、提出具体的预警警限的界定方法；最后，利用我国的统计资料，对ANN预警和粗集-神经网络相结合预警方法进行了实证分析和比较研究。
[Abstract]:This paper first summarizes the domestic and foreign macroeconomic early warning, puts forward the main problems existing in the macroeconomic early warning at present, and then expounds the theoretical system of macroeconomic early warning, and discusses the design of the early warning index. After analyzing and discussing the complementary advantages of the artificial neural network (ANN) early warning method and the rough set (RS) theory method, a macroeconomic early warning system model based on the combination of rough set and neural network is established. This paper discusses in detail the basic principle of early warning method based on rough set and neural network, and studies the realization of early warning method, including establishing the structure mode of rough set neural network early warning and solving the problems in the operation of rough set neural network early warning. Finally, using the statistical data of our country, the paper makes an empirical analysis and comparative study on the combination of ANN early warning and rough set neural network.
【学位授予单位】：河海大学
【学位级别】：硕士
【学位授予年份】：2003
【分类号】：F015
本文编号：2275881
[Abstract]:This paper first summarizes the domestic and foreign macroeconomic early warning, puts forward the main problems existing in the macroeconomic early warning at present, and then expounds the theoretical system of macroeconomic early warning, and discusses the design of the early warning index. After analyzing and discussing the complementary advantages of the artificial neural network (ANN) early warning method and the rough set (RS) theory method, a macroeconomic early warning system model based on the combination of rough set and neural network is established. This paper discusses in detail the basic principle of early warning method based on rough set and neural network, and studies the realization of early warning method, including establishing the structure mode of rough set neural network early warning and solving the problems in the operation of rough set neural network early warning. Finally, using the statistical data of our country, the paper makes an empirical analysis and comparative study on the combination of ANN early warning and rough set neural network.
【学位授予单位】：河海大学
【学位级别】：硕士
【学位授予年份】：2003
【分类号】：F015
【引证文献】
相关期刊论文 前4条
1 林健;彭敏晶;;基于主动学习的SVDD预警技术[J];辽宁工程技术大学学报;2006年S1期
2 朱勇;吴涛;;基于Rough集和构造性学习神经网络的经济预警模型[J];合肥工业大学学报(自然科学版);2007年07期
3 庞秀丽;冯玉强;庞志贤;;基于最大熵的经济预警研究[J];计算机工程与应用;2007年05期
4 汤小莉;逯颖;;浅析政府审计维护国家经济安全的主要作用及实现途径[J];未来与发展;2011年11期
相关硕士学位论文 前3条
1 刘金霞;地方财政风险及预警研究[D];河北工业大学;2005年
2 支小军;新疆生产建设兵团经济预警研究[D];石河子大学;2007年
3 夏璐;宏观经济预警方法研究-Kohonen-Bp神经网络模型[D];浙江工商大学;2010年
本文编号：2275881
本文链接：https://www.wllwen.com/jingjilunwen/jingjililun/2275881.html
