A Multi-scale Spectral Mapping Based Method to Identify the Significant Features of Funds' Investment Styles

›› 2018, Vol. 30 ›› Issue (3): 3-14.

A Multi-scale Spectral Mapping Based Method to Identify the Significant Features of Funds' Investment Styles

Sumuya^1,2, Yang Xiaoguang², Guo Chonghui³

1. School of Economics and Management, Inner Mongolia University, Hohhot 010021;
2. Academy of Mathematics and Systems Science, CAS, Beijing 100190;
3. Faculty of Management and Economics, Dalian University of Technology, Dalian 116024

Received:2016-12-27 Online:2018-03-28 Published:2018-03-26

Abstract

Abstract:

Investment style drift is a phenomenon that often occurs in securities investment funds, so identifying the significant features of their investment styles is important for selecting and rating them. Based on spectral mapping idea, we propose a multi-scale spectral mapping based method to identify the significant features of funds' investment styles and then logically expound basic principle of the method. Numerical simulation shows that multi-scale spectral mapping is a much more effective method than traditional ones to identify the significant features of funds' investment styles. In this paper this method is used to identify the significant features of the key investment styles that 24 Chinese open-end funds have taken on since 2012 and the result shows that there exist obvious convergence and drift in their investment styles.

Key words: multi-scale spectral mapping, Gaussian kernel parameter, significant features of investment style

Sumuya, Yang Xiaoguang, Guo Chonghui. A Multi-scale Spectral Mapping Based Method to Identify the Significant Features of Funds' Investment Styles[J]. , 2018, 30(3): 3-14.

References

[1] Sharpe W. Asset Allocation:Management Style and Performance Measurement[J]. Journal of Portfolio Management, 1992,18(2):7-19
[2] 彭耿. 基金投资风格漂移识别——基于收益和风险双维度[J]. 财经理论与实践, 2014,35(3):37-43
[3] Brookfield D., Su C., Bangassa K. Investment Style Positioning of UK Unit Trusts[J]. European Journal of Finance, 2015,21(10-11):946-970
[4] 孟庆斌,吴卫星,于上尧. 基金经理职业忧虑与其投资风格[J]. 经济研究, 2015,50(3):115-130
[5] 肖继辉,彭文平. 基金经理特征与投资能力、投资风格的关系[J]. 管理评论, 2012,24(7):40-48
[6] 肖继辉. 基金行业锦标赛及其激励效应研究——来自开放式基金的经验证据[J]. 南开管理评论, 2012,15(5):44-55
[7] 肖继辉,彭文平. 锦标赛制度与基金风险调整:理论拓展与经验证据[J]. 管理科学学报, 2015,18(1):87-98
[8] 骆盈盈,任颋. 基金管理模式、投资风格与经营绩效:来自中国公募基金探索性研究[J]. 现代财经, 2015,35(4):22-33
[9] 钱淑芳,许林,吴栩. 宏观经济环境对股票型基金投资风格漂移的影响研究[J]. 投资研究, 2015,9(4):33-45
[10] Wahal S., Yavuz M. Style Investing, Comovement and Return Predictability[J]. Journal of Financial Economics, 2013,107(1):136-154
[11] 易力,胡振华. 风格择时能力对基金绩效的影响研究[J]. 管理评论, 2016,28(4):41-51
[12] 刘莎莎,刘玉珍,唐涯. 信息优势、风险调整与基金业绩[J]. 管理世界, 2013,(8):67-76
[13] 郭文伟. 开放式基金投资风格漂移及风格资产轮换策略研究[D]. 上海交通大学博士学位论文, 2010
[14] Brown S., Goetzmann W. Mutual Fund Style[J]. Journal of Financial Economics, 1997,43(3):373-393
[15] Pattarin F., Paterlini S., Minerva T. Clustering Financial Time Series:An Application to Mutual Funds Style Analysis[J]. Computational Statistics Data Analysis, 2004,47(22):353-372
[16] 杨朝军,蔡明超,徐慧泉. 中国证券投资基金风格分类研究[J]. 上海交通大学学报, 2004,38(3):359-361
[17] 董铁牛,杨乃定,邵予工. 中国开放式基金投资风格分析[J]. 管理评论, 2008,20(7):3-9
[18] Fiedler M. Algebraic Connectivity of Graphs[J]. Czechoslovak Mathematical Journal, 1973,23(2):298-305
[19] Chung F. Spectral Graph Theory[M]. New York:American Mathematical Press, 1997
[20] Luxburg U. A Tutorial on Spectral Clustering[J]. Statistics and Computing, 2007,17(4):395-416
[21] Ng A., Jordan M., Weiss Y. On Spectral Clustering:Analysis and an Algorithm[C]//Advances in Neural Information Processing Systems. Cambridge:MIT Press, 2001
[22] Borjigin S., Guo C. Non-unique Cluster Numbers Determination Methods Based on Stability in Spectral Clustering[J]. Knowledge and Information Systems, 2013,36(2):439-458
[23] Shi J., Malik J. Normalized Cuts and Image Segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000,22(8):888-905
[24] Meila M., Shi J. Learning Segmentation by Random Walks[C]//Advances in Neural Information Processing Systems. Cambridge:MIT Press, 2001