HCGM-NET: A deep unfolding network for financial index tracking

Ruben Alan J Pauwels; Evangelia Tsiligianni; Nikos Deligiannis

HCGM-NET: A deep unfolding network for financial index tracking

Ruben Alan J Pauwels, Evangelia Tsiligianni, Nikos Deligiannis

Research output: Chapter in Book/Report/Conference proceeding › Conference paper

Abstract

Tracking the performance of a financial index by selecting asubset of assets composing the index is a problem that raisesseveral difficulties due to the large size of the stock market.Typically, optimisation algorithms with high complexity areemployed to address such problems. In this paper, we focus on sparse index tracking and employ a Frank-Wolfe-based algorithm which we translate into a deep neural network. This strategy, known as deep unfolding, leads to a learned model with high accuracy at a low computational cost. To the bestof our knowledge, this is the first deep unfolding design pro-posed for financial data processing. Numerical experimentsdemonstrate the superior performance of our approach.

Original language	English
Title of host publication	IEEE International Conference on Acoustics, Speech and Signal Processing
Subtitle of host publication	ICASSP
Publisher	IEEE
Pages	1-5
Number of pages	5
Publication status	Accepted/In press - 2021
Event	2021 IEEE International Conference on Acoustics, Speech and Signal Processing - Metro Toronto Convention Centre, Toronto, Canada Duration: 6 Jun 2021 → 11 Jun 2021 https://2021.ieeeicassp.org

Conference

Conference	2021 IEEE International Conference on Acoustics, Speech and Signal Processing
Abbreviated title	ICASSP 2021
Country/Territory	Canada
City	Toronto
Period	6/06/21 → 11/06/21
Internet address	https://2021.ieeeicassp.org

Keywords

financial index tracking
sparse portfolio selection
conditional gradient method
deep unfolding

SRP11: Strategic Research Programme: Processing of large scale multi-dimensional, multi-spectral, multi-sensorial and distributed data (M³D²)
Schelkens, P., Deligiannis, N., Jansen, B., Kuijk, M., Munteanu, A., Sahli, H., Steenhaut, K., Stiens, J., Schelkens, P., Cornelis, J. P., Kuijk, M., Munteanu, A., Sahli, H., Stiens, J. & Vounckx, R.
1/11/12 → 31/12/23
Project: Fundamental

Cite this

@inproceedings{1d664197c9e6475f900bc66984102c86,

title = "HCGM-NET: A deep unfolding network for financial index tracking",

abstract = "Tracking the performance of a financial index by selecting asubset of assets composing the index is a problem that raisesseveral difficulties due to the large size of the stock market.Typically, optimisation algorithms with high complexity areemployed to address such problems. In this paper, we focus on sparse index tracking and employ a Frank-Wolfe-based algorithm which we translate into a deep neural network. This strategy, known as deep unfolding, leads to a learned model with high accuracy at a low computational cost. To the bestof our knowledge, this is the first deep unfolding design pro-posed for financial data processing. Numerical experimentsdemonstrate the superior performance of our approach.",

keywords = "financial index tracking, sparse portfolio selection, conditional gradient method, deep unfolding",

author = "Pauwels, {Ruben Alan J} and Evangelia Tsiligianni and Nikos Deligiannis",

year = "2021",

language = "English",

pages = "1--5",

booktitle = "IEEE International Conference on Acoustics, Speech and Signal Processing",

publisher = "IEEE",

note = "2021 IEEE International Conference on Acoustics, Speech and Signal Processing , ICASSP 2021 ; Conference date: 06-06-2021 Through 11-06-2021",

url = "https://2021.ieeeicassp.org",

}

TY - GEN

T1 - HCGM-NET: A deep unfolding network for financial index tracking

AU - Pauwels, Ruben Alan J

AU - Tsiligianni, Evangelia

AU - Deligiannis, Nikos

PY - 2021

Y1 - 2021

N2 - Tracking the performance of a financial index by selecting asubset of assets composing the index is a problem that raisesseveral difficulties due to the large size of the stock market.Typically, optimisation algorithms with high complexity areemployed to address such problems. In this paper, we focus on sparse index tracking and employ a Frank-Wolfe-based algorithm which we translate into a deep neural network. This strategy, known as deep unfolding, leads to a learned model with high accuracy at a low computational cost. To the bestof our knowledge, this is the first deep unfolding design pro-posed for financial data processing. Numerical experimentsdemonstrate the superior performance of our approach.

AB - Tracking the performance of a financial index by selecting asubset of assets composing the index is a problem that raisesseveral difficulties due to the large size of the stock market.Typically, optimisation algorithms with high complexity areemployed to address such problems. In this paper, we focus on sparse index tracking and employ a Frank-Wolfe-based algorithm which we translate into a deep neural network. This strategy, known as deep unfolding, leads to a learned model with high accuracy at a low computational cost. To the bestof our knowledge, this is the first deep unfolding design pro-posed for financial data processing. Numerical experimentsdemonstrate the superior performance of our approach.

KW - financial index tracking

KW - sparse portfolio selection

KW - conditional gradient method

KW - deep unfolding

M3 - Conference paper

SP - 1

EP - 5

BT - IEEE International Conference on Acoustics, Speech and Signal Processing

PB - IEEE

T2 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing

Y2 - 6 June 2021 through 11 June 2021

ER -

HCGM-NET: A deep unfolding network for financial index tracking

Abstract

Conference

Keywords

Fingerprint

Projects

SRP11: Strategic Research Programme: Processing of large scale multi-dimensional, multi-spectral, multi-sensorial and distributed data (M³D²)

Cite this