Answers are grounded in this paper and cite their sources.
There is a growing appetite for parametric Japanese earthquake insurance products that use Japanese Meteorological Agency seismic intensity, IJMA, triggers. However, most earthquake risk models used to price these contracts do not compute IJMA directly, instead relying on conversions that introduce additional basis risk. Here, we present empirical parameterisations of IJMA as a function of all combinations of moment magnitude, hypocentral distance, peak ground acceleration (PGA), and spectral acceleration (Sa(T)) at 0.2, 0.3, 0.6, 1.0, 2.0, and 3.0 s, derived from earthquake risk model outputs. Parameterisations are fitted using ridge regression applied to 21,354 ground motion recordings from 112 Japanese earthquakes spanning January 1997 to December 2025. A parsimonious three-parameter model — combining PGA, Sa(0.3 s), and Sa(1.0 s) — outperforms a previously published parameterisation, substantially reducing both uncertainty and bias. When applied to a synthetic parametric insurance portfolio exposed to 25 years of Japanese seismicity, this parameterisation yields a measurable reduction in basis risk. These results support more reliable and transparent pricing of parametric earthquake insurance products.
No takes yet. Share an insight, caveat, or question.
Ball et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: