Low Financial Risk of Default and Productive Use of Assets Through Hidden Markov Models

Alexander Haro (), Genaro Sandoval, María Rodríguez, Victor Armijo, Ivonne Arana, William Vasquez, Elizabeth Proaño and Amanda Martínez
Additional contact information
Alexander Haro: Instituto Superior Tecnológico España, Unidad de Ciencias Empresariales, Ambato 180150, Ecuador
Genaro Sandoval: Facultad Ciencias Empresariales, Campus Callao, Universidad Cesar Vallejo, Lima 15314, Peru
María Rodríguez: Facultad de Ciencias de Gestión y Comunicaciones, Universidad Autónoma del Perú, Lima 15439, Peru
Victor Armijo: Facultad Ciencias Empresariales, Campus Callao, Universidad Cesar Vallejo, Lima 15314, Peru
Ivonne Arana: Facultad Ciencias Empresariales, Campus Callao, Universidad Cesar Vallejo, Lima 15314, Peru
William Vasquez: Facultad Ciencias Empresariales, Campus Callao, Universidad Cesar Vallejo, Lima 15314, Peru
Elizabeth Proaño: Instituto Superior Tecnológico España, Unidad de Ciencias Empresariales, Ambato 180150, Ecuador
Amanda Martínez: Facultad de Ciencias Económicas Administrativas y de Negocios, Universidad Tecnológica Indoamérica, Ambato 180101, Ecuador

Risks, 2025, vol. 13, issue 12, 1-20

Abstract: This paper analyzes solvency dynamics in Ecuador’s mutualist segment by modeling the joint behavior of the productive-assets-to-total-assets ratio (PATR) and portfolio-specific delinquency rates. Using monthly supervisory data from the Superintendencia de Economía Popular y Solidaria (SEPS) for the full universe of four mutualist institutions (2022–2025), we estimate a multivariate Gaussian Hidden Markov Model on system-level aggregates. The model identifies latent regimes that summarize configurations of asset productivity and segmented credit risk, distinguishing relatively sound conditions from episodes of heightened vulnerability. Model selection is based on information criteria, complemented by convergence checks, distributional diagnostics, and alternative covariance specifications to assess robustness. The approach is explicitly framed as diagnostic rather than causal or prescriptive: it does not replace simple thresholds nor calibrate capital buffers, but organizes supervisory information into interpretable solvency states with associated frequencies and expected durations. The framework is transparent and reproducible and provides a baseline for future extensions with longer samples and richer covariates.

Keywords: bank solvency; delinquency; hidden Markov models; financial risk; financial system; finance; machine learning (search for similar items in EconPapers)
JEL-codes: C G0 G1 G2 G3 K2 M2 M4 (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2227-9091/13/12/230/pdf (application/pdf)
https://www.mdpi.com/2227-9091/13/12/230/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jrisks:v:13:y:2025:i:12:p:230-:d:1804457

Access Statistics for this article

Risks is currently edited by Mr. Claude Zhang

More articles in Risks from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().