Regression-Based Index Tracking Versus Clustering-Based Index Tracking: An Empirical Study

Fangyu Zhang; Qintong Lyu; Jun Wang

doi:10.1007/978-981-95-1233-1_5

Regression-Based Index Tracking Versus Clustering-Based Index Tracking: An Empirical Study

Fangyu Zhang^*
, Qintong Lyu
, Jun Wang

^*Corresponding author for this work

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

1 Citation (Scopus)

Abstract

In this paper, regression-based and clustering-based index tracking methods are compared in terms of tracking accuracy, solution consistency, portfolio volatility, and downside risk. The former is based on least-squares regression under a cardinality constraint. The latter is based on K-means, K-medoids, and hierarchical clustering algorithms with dissimilarity metrics defined on Euclidean distance, Pearson correlation coefficient, and dynamic time warping. Experimental results on major world stock markets show that the regression-based method significantly outperforms the clustering-based methods in terms of tracking accuracy and consistency, while the index tracking method based on hierarchical clustering and Pearson correlation coefficient results in slightly lower volatility and downside risk than the regression-based method. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026.

Original language	English
Title of host publication	Advances in Neural Networks - ISNN 2025
Subtitle of host publication	19th International Symposium on Neural Networks, Proceedings
Editors	Long Jin, Lidan Wang
Publisher	Springer Singapore
Pages	51-62
Number of pages	12
Edition	1
ISBN (Electronic)	978-981-95-1233-1
ISBN (Print)	978-981-95-1232-4
DOIs	https://doi.org/10.1007/978-981-95-1233-1_5
Publication status	Published - 23 Aug 2025
Event	19th International Symposium on Neural Networks, ISNN 2025 - Zhangye, China Duration: 22 Aug 2025 → 24 Aug 2025

Publication series

Name	Lecture Notes in Computer Science
Volume	15951 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	19th International Symposium on Neural Networks, ISNN 2025
Place	China
City	Zhangye
Period	22/08/25 → 24/08/25

Funding

This work was supported in part by the Research Grants Council of the Hong Kong Special Administrative Region of China (Grant 11202019), in part by the InnoHK initiative of the Innovation and Technology Commission of the Hong Kong Special Administrative Region Government, and in part by the Laboratory for AI-Powered Financial Technologies, Hong Kong.

Research Keywords

Cardinality constraint
Clustering
Financial index tracking
Least-squares regression
Portfolio selection

RGC Funding Information

RGC-funded

Access Output

10.1007/978-981-95-1233-1_5

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

Zhang, F., Lyu, Q., & Wang, J. (2025). Regression-Based Index Tracking Versus Clustering-Based Index Tracking: An Empirical Study. In L. Jin, & L. Wang (Eds.), Advances in Neural Networks - ISNN 2025: 19th International Symposium on Neural Networks, Proceedings (1 ed., pp. 51-62). (Lecture Notes in Computer Science; Vol. 15951 LNCS). Springer Singapore. https://doi.org/10.1007/978-981-95-1233-1_5

@inproceedings{8095c250ac5e4059843a3d1b21f53066,

title = "Regression-Based Index Tracking Versus Clustering-Based Index Tracking: An Empirical Study",

abstract = "In this paper, regression-based and clustering-based index tracking methods are compared in terms of tracking accuracy, solution consistency, portfolio volatility, and downside risk. The former is based on least-squares regression under a cardinality constraint. The latter is based on K-means, K-medoids, and hierarchical clustering algorithms with dissimilarity metrics defined on Euclidean distance, Pearson correlation coefficient, and dynamic time warping. Experimental results on major world stock markets show that the regression-based method significantly outperforms the clustering-based methods in terms of tracking accuracy and consistency, while the index tracking method based on hierarchical clustering and Pearson correlation coefficient results in slightly lower volatility and downside risk than the regression-based method. {\textcopyright} The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026.",

keywords = "Cardinality constraint, Clustering, Financial index tracking, Least-squares regression, Portfolio selection",

author = "Fangyu Zhang and Qintong Lyu and Jun Wang",

year = "2025",

month = aug,

day = "23",

doi = "10.1007/978-981-95-1233-1\_5",

language = "English",

isbn = "978-981-95-1232-4",

series = "Lecture Notes in Computer Science",

publisher = "Springer Singapore",

pages = "51--62",

editor = "Long Jin and Lidan Wang",

booktitle = "Advances in Neural Networks - ISNN 2025",

address = "Singapore",

edition = "1",

note = "19th International Symposium on Neural Networks, ISNN 2025 ; Conference date: 22-08-2025 Through 24-08-2025",

}

Zhang, F , Lyu, Q & Wang, J 2025, Regression-Based Index Tracking Versus Clustering-Based Index Tracking: An Empirical Study. in L Jin & L Wang (eds), Advances in Neural Networks - ISNN 2025: 19th International Symposium on Neural Networks, Proceedings. 1 edn, Lecture Notes in Computer Science, vol. 15951 LNCS, Springer Singapore, pp. 51-62, 19th International Symposium on Neural Networks, ISNN 2025, Zhangye, China, 22/08/25. https://doi.org/10.1007/978-981-95-1233-1_5

Regression-Based Index Tracking Versus Clustering-Based Index Tracking: An Empirical Study. / Zhang, Fangyu ; Lyu, Qintong ; Wang, Jun.
Advances in Neural Networks - ISNN 2025: 19th International Symposium on Neural Networks, Proceedings. ed. / Long Jin; Lidan Wang. 1. ed. Springer Singapore, 2025. p. 51-62 (Lecture Notes in Computer Science; Vol. 15951 LNCS).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Regression-Based Index Tracking Versus Clustering-Based Index Tracking

T2 - 19th International Symposium on Neural Networks, ISNN 2025

AU - Zhang, Fangyu

AU - Lyu, Qintong

AU - Wang, Jun

PY - 2025/8/23

Y1 - 2025/8/23

N2 - In this paper, regression-based and clustering-based index tracking methods are compared in terms of tracking accuracy, solution consistency, portfolio volatility, and downside risk. The former is based on least-squares regression under a cardinality constraint. The latter is based on K-means, K-medoids, and hierarchical clustering algorithms with dissimilarity metrics defined on Euclidean distance, Pearson correlation coefficient, and dynamic time warping. Experimental results on major world stock markets show that the regression-based method significantly outperforms the clustering-based methods in terms of tracking accuracy and consistency, while the index tracking method based on hierarchical clustering and Pearson correlation coefficient results in slightly lower volatility and downside risk than the regression-based method. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026.

AB - In this paper, regression-based and clustering-based index tracking methods are compared in terms of tracking accuracy, solution consistency, portfolio volatility, and downside risk. The former is based on least-squares regression under a cardinality constraint. The latter is based on K-means, K-medoids, and hierarchical clustering algorithms with dissimilarity metrics defined on Euclidean distance, Pearson correlation coefficient, and dynamic time warping. Experimental results on major world stock markets show that the regression-based method significantly outperforms the clustering-based methods in terms of tracking accuracy and consistency, while the index tracking method based on hierarchical clustering and Pearson correlation coefficient results in slightly lower volatility and downside risk than the regression-based method. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026.

KW - Cardinality constraint

KW - Clustering

KW - Financial index tracking

KW - Least-squares regression

KW - Portfolio selection

UR - https://www.scopus.com/pages/publications/105014494849

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-105014494849&origin=recordpage

U2 - 10.1007/978-981-95-1233-1_5

DO - 10.1007/978-981-95-1233-1_5

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 978-981-95-1232-4

T3 - Lecture Notes in Computer Science

SP - 51

EP - 62

BT - Advances in Neural Networks - ISNN 2025

A2 - Jin, Long

A2 - Wang, Lidan

PB - Springer Singapore

Y2 - 22 August 2025 through 24 August 2025

ER -

Zhang F , Lyu Q , Wang J. Regression-Based Index Tracking Versus Clustering-Based Index Tracking: An Empirical Study. In Jin L, Wang L, editors, Advances in Neural Networks - ISNN 2025: 19th International Symposium on Neural Networks, Proceedings. 1 ed. Springer Singapore. 2025. p. 51-62. (Lecture Notes in Computer Science). Epub 2025 Aug 19. doi: 10.1007/978-981-95-1233-1_5

Regression-Based Index Tracking Versus Clustering-Based Index Tracking: An Empirical Study

Abstract

Publication series

Conference

Funding

Research Keywords

RGC Funding Information

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

GRF: Collaborative Neurodynamic Approaches to Portfolio Optimization

Cite this

Regression-Based Index Tracking Versus Clustering-Based Index Tracking: An Empirical Study

Abstract

Publication series

Conference

Funding

Research Keywords

RGC Funding Information

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Projects

GRF: Collaborative Neurodynamic Approaches to Portfolio Optimization

Cite this