EDS researchers Juan Juan Cai, Yicong Lin, Julia Schaumburg and Chenhui Wang publish a new study introducing a nonparametric framework for measuring the persistence of extreme events. The method is used to investigate how persistent extreme heat tends to be and how its probability of occurrence has changed over time.
The study proposes a nonparametric framework for estimating the extremal index, which captures the persistence of extreme observations. The approach is applied to summer apparent temperature data from nine European cities covering the period 1940–2025.
The results reveal strong persistence in extreme temperatures across all cities, with extreme heat episodes typically lasting at least two days. Moreover, the analysis shows a marked increase in the occurrence of persistent heat events over time: the probability of two-day extreme-temperature episodes is estimated to be two to four times higher in the most recent three decades than in the period 1940–1974.
The article titled "Estimation and inference for the persistence of extremely high temperatures” and is available online.
The Econometrics Journal is a leading international journal that publishes high-quality research on econometric theory, methods, and applications.