Explainable least square Monte Carlo for solvency capital requirement evaluation
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<rdf:Description>
<dc:creator>Perla, Francesca</dc:creator>
<dc:date>2026-03-16</dc:date>
<dc:description xml:lang="es">Sumario: The article presents an explainable extension of the Least Squares Monte Carlo (LSMC) method for estimating the Solvency Capital Requirement (SCR) within the Solvency II framework. The proposed approach combines Monte Carlo simulation techniques with explainable deep learning models, specifically the localGLMnet architecture, to reduce the computational cost of nested simulations. This approach makes it possible to maintain accuracy in SCR estimation while simultaneously improving model interpretability. The results are validated through numerical experiments on realistic life insurance portfolios. In addition, the use of ElasticNet regularization is analyzed to identify the most relevant risk factors</dc:description>
<dc:identifier>https://documentacion.fundacionmapfre.org/documentacion/publico/es/bib/190505.do</dc:identifier>
<dc:language>spa</dc:language>
<dc:rights xml:lang="es">InC - http://rightsstatements.org/vocab/InC/1.0/</dc:rights>
<dc:subject xml:lang="es">Solvencia II</dc:subject>
<dc:subject xml:lang="es">Simulación Monte Carlo</dc:subject>
<dc:subject xml:lang="es">Cálculo actuarial</dc:subject>
<dc:subject xml:lang="es">Riesgo financiero</dc:subject>
<dc:subject xml:lang="es">Seguro de vida</dc:subject>
<dc:subject xml:lang="es">Requerimientos financieros</dc:subject>
<dc:type xml:lang="es">Artículos y capítulos</dc:type>
<dc:title xml:lang="es">Explainable least square Monte Carlo for solvency capital requirement evaluation</dc:title>
<dc:relation xml:lang="es">En: North American actuarial journal. - Schaumburg : Society of Actuaries, 1997- = ISSN 1092-0277. - 16/03/2026 Tomo 30 Número 1 - 2026 , 24 p.</dc:relation>
</rdf:Description>
</rdf:RDF>