Topology of products similarity network for market forecasting

Jingfang FAN, Keren COHEN, Louis M. SHEKHTMAN, Sibo LIU, Jun MENG, Yoram LOUZOUN, Shlomo HAVLIN

Research output: Journal PublicationsJournal Article (refereed)

Abstract

The detection and prediction of risk in financial markets is one of the main challenges of economic forecasting, and draws much attention from the scientific community. An even more challenging task is the prediction of the future relative gain of companies. We here develop a novel combination of product text analysis, network theory and topological based machine learning to study the future performance of companies in financial markets. Our network links are based on the similarity of firms’ products and constructed using the Securities Exchange Commission (SEC) filings of US listed firms. We find that several topological features of this network can serve as good precursors of risks or future gain of companies. We then apply machine learning to network attributes vectors for each node to predict successful and failing firms. The resulting accuracies are much better than current state of the art techniques. The framework presented here not only facilitates the prediction of financial markets but also provides insight and demonstrates the power of combining network theory and topology based machine learning.
Original languageEnglish
Article number69
JournalApplied Network Science
Volume4
Issue number1
Early online date28 Aug 2019
DOIs
Publication statusPublished - Dec 2019

Fingerprint

Forecasting
Topology
Learning systems
Financial Markets
Circuit theory
Machine Learning
Prediction
Industry
Text Analysis
Precursor
Economics
Attribute
Similarity
Market
Predict
Financial markets
Vertex of a graph
Demonstrate
Business

Keywords

  • Economic
  • Machine learning
  • Network
  • Topology

Cite this

FAN, J., COHEN, K., SHEKHTMAN, L. M., LIU, S., MENG, J., LOUZOUN, Y., & HAVLIN, S. (2019). Topology of products similarity network for market forecasting. Applied Network Science, 4(1), [69]. https://doi.org/10.1007/s41109-019-0171-y
FAN, Jingfang ; COHEN, Keren ; SHEKHTMAN, Louis M. ; LIU, Sibo ; MENG, Jun ; LOUZOUN, Yoram ; HAVLIN, Shlomo. / Topology of products similarity network for market forecasting. In: Applied Network Science. 2019 ; Vol. 4, No. 1.
@article{8143e37595a64f29bf75328e77ef13ad,
title = "Topology of products similarity network for market forecasting",
abstract = "The detection and prediction of risk in financial markets is one of the main challenges of economic forecasting, and draws much attention from the scientific community. An even more challenging task is the prediction of the future relative gain of companies. We here develop a novel combination of product text analysis, network theory and topological based machine learning to study the future performance of companies in financial markets. Our network links are based on the similarity of firms’ products and constructed using the Securities Exchange Commission (SEC) filings of US listed firms. We find that several topological features of this network can serve as good precursors of risks or future gain of companies. We then apply machine learning to network attributes vectors for each node to predict successful and failing firms. The resulting accuracies are much better than current state of the art techniques. The framework presented here not only facilitates the prediction of financial markets but also provides insight and demonstrates the power of combining network theory and topology based machine learning.",
keywords = "Economic, Machine learning, Network, Topology",
author = "Jingfang FAN and Keren COHEN and SHEKHTMAN, {Louis M.} and Sibo LIU and Jun MENG and Yoram LOUZOUN and Shlomo HAVLIN",
year = "2019",
month = "12",
doi = "10.1007/s41109-019-0171-y",
language = "English",
volume = "4",
journal = "Applied Network Science",
issn = "2364-8228",
publisher = "Springer Open",
number = "1",

}

FAN, J, COHEN, K, SHEKHTMAN, LM, LIU, S, MENG, J, LOUZOUN, Y & HAVLIN, S 2019, 'Topology of products similarity network for market forecasting', Applied Network Science, vol. 4, no. 1, 69. https://doi.org/10.1007/s41109-019-0171-y

Topology of products similarity network for market forecasting. / FAN, Jingfang; COHEN, Keren; SHEKHTMAN, Louis M.; LIU, Sibo; MENG, Jun; LOUZOUN, Yoram; HAVLIN, Shlomo.

In: Applied Network Science, Vol. 4, No. 1, 69, 12.2019.

Research output: Journal PublicationsJournal Article (refereed)

TY - JOUR

T1 - Topology of products similarity network for market forecasting

AU - FAN, Jingfang

AU - COHEN, Keren

AU - SHEKHTMAN, Louis M.

AU - LIU, Sibo

AU - MENG, Jun

AU - LOUZOUN, Yoram

AU - HAVLIN, Shlomo

PY - 2019/12

Y1 - 2019/12

N2 - The detection and prediction of risk in financial markets is one of the main challenges of economic forecasting, and draws much attention from the scientific community. An even more challenging task is the prediction of the future relative gain of companies. We here develop a novel combination of product text analysis, network theory and topological based machine learning to study the future performance of companies in financial markets. Our network links are based on the similarity of firms’ products and constructed using the Securities Exchange Commission (SEC) filings of US listed firms. We find that several topological features of this network can serve as good precursors of risks or future gain of companies. We then apply machine learning to network attributes vectors for each node to predict successful and failing firms. The resulting accuracies are much better than current state of the art techniques. The framework presented here not only facilitates the prediction of financial markets but also provides insight and demonstrates the power of combining network theory and topology based machine learning.

AB - The detection and prediction of risk in financial markets is one of the main challenges of economic forecasting, and draws much attention from the scientific community. An even more challenging task is the prediction of the future relative gain of companies. We here develop a novel combination of product text analysis, network theory and topological based machine learning to study the future performance of companies in financial markets. Our network links are based on the similarity of firms’ products and constructed using the Securities Exchange Commission (SEC) filings of US listed firms. We find that several topological features of this network can serve as good precursors of risks or future gain of companies. We then apply machine learning to network attributes vectors for each node to predict successful and failing firms. The resulting accuracies are much better than current state of the art techniques. The framework presented here not only facilitates the prediction of financial markets but also provides insight and demonstrates the power of combining network theory and topology based machine learning.

KW - Economic

KW - Machine learning

KW - Network

KW - Topology

UR - http://www.scopus.com/inward/record.url?scp=85071561422&partnerID=8YFLogxK

U2 - 10.1007/s41109-019-0171-y

DO - 10.1007/s41109-019-0171-y

M3 - Journal Article (refereed)

VL - 4

JO - Applied Network Science

JF - Applied Network Science

SN - 2364-8228

IS - 1

M1 - 69

ER -