PERSEUS - Prediction of extreme events with reliable uncertainty quantification and risk assessment

Italian Ministry of Education, University and Research; Italia; On-site; EUR 25,000 - 40,000; Full time

Job summary

Università degli Studi di Padova - Dipartimento di Scienze Statistiche is seeking researchers to contribute to the PERSEUS project, advancing extreme value analysis and Bayesian inference. The work encompasses developing novel tools to infer dependency structures among variables and building probabilistic forecasting methods for extremes, including time-dependent scenarios and cascading effects.

The fellowship targets researchers within Europe, offering engagement in innovative statistical

Qualifications

• Eligibility of fellows: country/ies of residence: EUROPE
• Eligibility of fellows: nationality/ies: EUROPE

Responsibilities

• PERSEUS aims at advancing extreme value analysis along two main intertwined directions: development of novel Bayesian tools and probabilistic forecasting methods for extremes.
• Focus areas include dependence structures among variables and predictive inference for extremes in time series and cascading effects.

Job description

Organisation/Company Università degli Studi di Padova - Dipartimento di Scienze Statistiche Research Field Mathematics Researcher Profile Recognised Researcher (R2) Leading Researcher (R4) First Stage Researcher (R1) Established Researcher (R3) Application Deadline 28 Oct 2026 - 13:00 (UTC) Country Italy Type of Contract To be defined Job Status Not Applicable Is the job funded through the EU Research Framework Programme? Not funded by a EU programme Is the Job related to staff position within a Research Infrastructure? No

Offer Description

PERSEUS aims at advancing extreme value analysis along two main intertwined directions: 1) development of novel Bayesian tools, based on posterior distributions, to infer the dependency structure among variables, underlying the occurrence of extreme events, and quantify related uncertainty. Dependence forms include
either extremal dependence between variables taking large values simultaneously, or the impact of non extreme values of certain variables on the tail of others, covering causal effects. 2) Construction of probabilistic forecasting methods, and techniques for predictive inference for extremes, while explicitly accounting for
dependence, such as serial dependence in the context of extreme value prediction over time, cascading effects, extremal regression structures, and treatment effects in extreme causal inference.

• Italy

Eligibility of fellows: country/ies of residence:

• EUROPE

Eligibility of fellows: nationality/ies: