基于指标独立性的景气指数构建赋权方法
发布时间：2018-09-04 14:21
【摘要】：经济景气指数常使用独立性权数来合成。以往的研究中,只使用一种相关系数作为指标间信息重复程度的代表,并且未能突显出指标的预测能力。文章提出了复合独立信息赋权法。其原理是使用复相关系数,时差相关系数,自相关系数,用时差进行修正,再进行归一化得到权重。以广东省经济数据为例,使用简单相关系数法、复相关系数法等常见的独立信息赋权方法进行了比较。认为复合独立信息赋权法具有能够突显具有优良预测能力指标的能力。
[Abstract]:The economic boom index often uses independence weights to synthesize. In previous studies, only one correlation coefficient was used as the representative of the degree of information repetition among indicators, and the predictive ability of the indicators was not highlighted. In this paper, a composite independent information weighting method is proposed. The principle is to use the complex correlation coefficient, the time difference correlation coefficient, the autocorrelation coefficient, the time difference correction, and then normalize the weight. Taking the economic data of Guangdong Province as an example, this paper compares the common independent information weighting methods such as simple correlation coefficient method and complex correlation coefficient method. It is considered that the composite independent information weighting method has the ability to highlight the indicators with good predictive ability.
【作者单位】： 中国南方电网科学研究院;厦门大学数据挖掘研究中心;
【基金】：中国南方电网有限责任公司电网技术研究中心资助项目
【分类号】：F127;F224
本文编号：2222409
[Abstract]:The economic boom index often uses independence weights to synthesize. In previous studies, only one correlation coefficient was used as the representative of the degree of information repetition among indicators, and the predictive ability of the indicators was not highlighted. In this paper, a composite independent information weighting method is proposed. The principle is to use the complex correlation coefficient, the time difference correlation coefficient, the autocorrelation coefficient, the time difference correction, and then normalize the weight. Taking the economic data of Guangdong Province as an example, this paper compares the common independent information weighting methods such as simple correlation coefficient method and complex correlation coefficient method. It is considered that the composite independent information weighting method has the ability to highlight the indicators with good predictive ability.
【作者单位】： 中国南方电网科学研究院;厦门大学数据挖掘研究中心;
【基金】：中国南方电网有限责任公司电网技术研究中心资助项目
【分类号】：F127;F224
【相似文献】
相关期刊论文 前8条
1 任静;;提高多指标决策客观性的赋权方法[J];管理评论;2012年05期
2 李明奇;刘玉娟;赵美玲;;一种基于判断矩阵的专家赋权方法[J];科技信息;2010年21期
3 申锦标;吕跃进;;一种基于向量贴近度的组合赋权方法[J];重庆工学院学报(自然科学版);2009年02期
4 张玉;魏华波;;基于CRITIC的多属性决策组合赋权方法[J];统计与决策;2012年16期
5 张华;王东明;王晶日;吕波;;建设节水型社会评价指标体系及赋权方法研究[J];环境保护科学;2010年05期
6 何成铭;李文鹏;冯靖;;基于DEA和G1法的指标综合赋权方法[J];四川兵工学报;2010年10期
7 朱建军;吴伟丽;刘思峰;;一种基于模糊线性规划的主客观组合赋权方法[J];运筹与管理;2006年03期
8 ;[J];;年期
相关硕士学位论文 前1条
1 马艳霞;土地评价指标赋权方法研究[D];甘肃农业大学;2011年
，本文编号：2222409
本文链接：https://www.wllwen.com/jingjilunwen/hongguanjingjilunwen/2222409.html
