Forecasting Financial Crisis using Topological Data Analysis Approach

Naomi Nyaboke Oseko; Achuo Gilead Omondi; Hassan Dogo Onyango; Desma Awuor Olwa; Gabriel Maina; Moses Oruru Morara; Kelvin Mwangi Thiong’o

doi:10.51867/Asarev.Maths.1.1.1

Authors

Naomi Nyaboke Oseko Department of Mathematics, Masinde Muliro University of Science and Technology, P. O. Box 190-50100, Kakamega, Kenya Author
Achuo Gilead Omondi Department of Mathematics, Masinde Muliro University of Science and Technology, P. O. Box 190-50100, Kakamega, Kenya Author
Hassan Dogo Onyango Department of Mathematics, Masinde Muliro University of Science and Technology, P. O. Box 190-50100, Kakamega, Kenya Author
Desma Awuor Olwa Department of Mathematics, Masinde Muliro University of Science and Technology, P. O. Box 190-50100, Kakamega, Kenya Author
Gabriel Maina Department of Mathematics, Masinde Muliro University of Science and Technology, P. O. Box 190-50100, Kakamega, Kenya Author
Moses Oruru Morara Department of Mathematics, Masinde Muliro University of Science and Technology, P. O. Box 190-50100, Kakamega, Kenya Author
Kelvin Mwangi Thiong’o Department of Mathematics, Masinde Muliro University of Science and Technology, P. O. Box 190-50100, Kakamega, Kenya Author

DOI:

https://doi.org/10.51867/Asarev.Maths.1.1.1

Keywords:

Financial Crisis, Topological Data Analysis, Traditional Financial Forecasting

Abstract

Traditional financial forecasting methods often struggle to capture the complex interactions and emerging patterns that precede financial crises. By leveraging on TDA, this research aims to uncover potential topological features that might serve as early warning signals for impending financial crises. The study adopts the utilization of Topological Data Analysis, an initiative mathematical framework to explore and analyze the inherent topological structures within financial data set, using secondary data from ”Yahoo Finance API”. The results of the analysis conducted using Python indicate that persistence homology in TDA successfully identifies key topology features associated with financial crises, implying its potential for developing early warning systems in the financial sector. The insights gained from this analysis could significantly enhance the early detection and proactive management of system risks in financial market, thereby contributing to more robust risk assessment and policy formulation strategies.

References

Andrew, M. and Vasileios, M. A. 2016 point process distance on the space of persistence diagrams. Submitted to Advances in Data Analysis and Classification, Sept. 2016. https://sites.google. comsiteandrewmarchesemath/journal_paper_pre.pdf?attredirects=0&d=1.

Bauer, U. (2017). A lean C++ code for the computation of Vietoris-Rips persistence barcodes. Software. available at. https://github.com/Ripser/ripser (Accessed February 16, 2024).

Carlos, M., Dans, F and Bien, C. (2009-2023). World Financial Crises: Lessons of the past(pages 34-35). https://doi.org/10.3917/fbc.034.0034 DOI: https://doi.org/10.3917/fbc.034.0034

Coudert, V and Idier, J., (2018). Reducing model risk in early warning systems for banking crises in the euro area, International Economics, Elsevier, vol.156(C), pages 98-116. https://doi.org/10.1016/j.inteco.2018.01.002 DOI: https://doi.org/10.1016/j.inteco.2018.01.002

Carlsson, G. (2009). Topology and data. Bulletin of the American Mathematical Society, 46(2), 255-308. https://doi.org/10.1090/S0273-0979-09-01249-X DOI: https://doi.org/10.1090/S0273-0979-09-01249-X

Chazal, F. (2017). High-dimensional topological data analysis. In Handbook of Discrete and Computational Geometry (3rd Ed - To appear), chapter 27. CRC Press.

Carlsson, G., Zomorodian, A. (2005). Computing persistent homology. Discrete Com put. Geom. 33, 249-274. https://doi.org/10.1007/s00454-004-1146-y

Edelsbrunner, H., Harer, J. (2009). Computational Topology: An Introduction. Amer. Math. Soc., Providence. https://doi.org/10.1090/mbk/069

Edelsbrunner, H., Harer, J. (2010). Persistent homology-a survey. Contemporary Mathematics, vol. 453.

Amer, M. Soc., 2008, 257-282.

Cris, V. (2023). Predicting Market Crashes With Topology in Python.Beyond Traditional Indicators for Insights Into Market Behavior. https://medium.com/ predicting-stock-market-crashes-with-topological-data-analysis-in-python.

Edelsbrunner, L. and Zomorodian (2002). Topological Persistence and simplification 28,511-533. https://doi.org/10.1007/s00454-002-2885-2

Edelsbrunner H., Letscher D., and Zomordian D.(2002). Topological persistence and simplification, Descrete Comput. Geom. (28) 511.

https://doi.org/10.1007/s00454-002-2885-2 DOI: https://doi.org/10.1007/s00454-002-2885-2

Edelsbrunner, H., and Harer, J.(2010). Computational topology: an introduction (Amer. Math. Soc.). https://doi.org/10.1090/mbk/069 DOI: https://doi.org/10.1090/mbk/069

Julien, I and Virginie, C.(2018) Predicting financial crises: a balancing act.

Gunnar, C. (2009) Topology and Data

Minsky, H. (2016). The financial Instability Hyponthesis: A Restatement.

Lum P.Y. ,Singh G., Lehman A., Ishkanov T., Vejdemo-Johansson M., Alagappan M, Carlsson J., and Carlsson G. (2013). Extracting insights from the shape of complex data using topology, Scientific Reports (3) 1236. https://doi.org/10.1038/srep01236 DOI: https://doi.org/10.1038/srep01236

Polterovich, L., et al. (2019). Topological persistence in geometry and analysis. arXiv: 1904.04044. https://doi.org/10.1090/ulect/074 DOI: https://doi.org/10.1090/ulect/074

Topology for Computing by Afra J. Zomorodian

Gunnar, C. (2005). Afra Zomorodian. Computing persistent homology. Discrete and Computational Geometry. https://geometry.stanford.edu/papers/zc-cph-05/zc-cph-05.pdf.

Salnikov V, Cassese D, Lambiotte R, Jones NS. (2018). Co-occurrence simplicial complexes in mathematics: identifying the holes of knowledge. Epub 2018/01/01. Appl Netw Sci 3(1):37. PubMed PMID: 30839828; PubMedCentral PMCID: PMCPMC6214324. https://doi.org/10.1007/s41109-018-0074-3 DOI: https://doi.org/10.1007/s41109-018-0074-3

Goh WP, Kwek D, Hogan D, Cheong S.,A. (2014). Complex network analysis of teaching practices.EPJ Data Science 3(1):1-16. https://doi.org/10.1140/epjds/s13688-014-0034-9 DOI: https://doi.org/10.1140/epjds/s13688-014-0034-9

Gidea M., and Katz Y. (2018). Topological data analysis of financial time series:Landscapes of crashes. Physica A: Statistical Mechanics and its Applications, 491:820 - 834. https://doi.org/10.1016/j.physa.2017.09.028 DOI: https://doi.org/10.1016/j.physa.2017.09.028

Chazal, F., Glisse, M., Labru'ere, C., and Michel, B. Convergence rates for persistence diagram estimation in topological data analysis. In Proceedings of the 31st International Conference on Machine Learning (ICML-14), pages 163-171

Peter, B.(2015). Statistical topological data analysis using persistence landscapes. Journal of Machine Learning Research, 16(1):77-102.

Ulrich, B. Ripser, (2015). https://github.com/Ripser/ripser

Otter N, Porter MA, Tillmann U, Grindrod P, Harrington HA (2017). A roadmap for the computation of persistent homology. Epub. EPJ Data Sci (2017) 6(1):17, 2017 . PubMed PMID: 32025466; PubMed Central PMCID: PMCPMC6979512. https://doi.org/10.1140/epjds/s13688-017-0109-5 DOI: https://doi.org/10.1140/epjds/s13688-017-0109-5

Hausmann J-C. On the Vietoris-Rips complexes and a cohomology theory for metric spaces. Ann Math Stud (1995) 138:175-88. https://doi.org/10.1515/9781400882588-013 DOI: https://doi.org/10.1515/9781400882588-013

Sean, R. (2021) Three financial crises in 21st Century(updated June 2023),https://www.investopedia.com/articles/investing/011116/3-financial-crises-21st-century.asp

Will, K. (2022) ,Financial Crisis:Definition,causes and examples(updated February 15,2023), https://www.investopedia.com/terms/f/financial-crisis.asp

Yeshwanth, N.(2023). Topological Data Analysis.

Zomorodian, A. J. (2005). Topology for computing (cambridge monographs on applied and computational mathematics) (1 ed.). Cambridge University Press.

Zomorodian, A. and Carlsson, G. (2005). Computing Persistent Homology 33, 249- 274.

https://doi.org/10.1007/s00454-004-1146-y DOI: https://doi.org/10.1007/s00454-004-1146-y