基于时态聚类模糊综合评价的基金评级算法  

Research on fund rating method based on fuzzy comprehensive evaluation with temporal clustering

作　　者：孟志青[1] 金诗思 MENG Zhiqing;JIN Shisi(School of Management,Zhejiang University of Technology,Hangzhou 310023,China)

机构地区：[1]浙江工业大学管理学院,浙江杭州310023

出　　处：《浙江工业大学学报》2021年第5期520-531,共12页Journal of Zhejiang University of Technology

基　　金：浙江省自然科学基金资助项目(LY18A010031)。

摘　　要：基金评级指标数据是一种具有时间属性的时态数据,被评定的基金级别是随着时间变化的,因此,如何给出一个客观的基金评价方法是非常困难的问题。通过对基金数据在时间粒度上的变换,改进了模糊综合评价法,建立了一种新的基于时态聚类模糊综合评价的基金评级算法(以下简称TCF算法)。针对科技、传媒及通讯类行业基金进行了实证分析,从准确性和稳定性两个角度表明:TCF算法具有良好的适用性,同时随着时间的变化,采用TCF算法所得到的评级结果比其他基金评级机构的评级结果更优,因此该算法对提高证券基金评级的准确性有一定的帮助,能为基金评价理论提供一种新的算法。Fund rating index data is a kind of temporal data with time attribute,and fund level assessed is changing with time,so it is very difficult to give an objective fund evaluation method.Through the transformation of fund data in time granularity,a new fund rating algorithm based on temporal clustering fuzzy comprehensive evaluation is established in this paper(hereinafter referred to as TCF algorithm).An empirical analysis on the funds of science and technology,media and communication industriesis implemented,and the result shows that the TCF algorithm has good applicabilityfrom the perspectives of accuracy and stability,respectively.At the same time,the rating results obtained by using TCF algorithm are better than those of other fund rating agencieswith the changing of time.Therefore,this algorithm can help to improve the accuracy of securities fund rating and provide a new algorithm for fund evaluation theory.

关 键 词：时态数据 基金评级 模糊综合评价 时态聚类 

分 类 号：F830.91[经济管理—金融学]