Beyond the Gale: Parametric Insurance Pricing for Offshore Wind: Applications of Hierarchical Bayesian Methods, Markov Chain Monte Carlo and Copula-Based Risk Assessment on Zero Generation Events

Abstract

This thesis introduces an innovative pricing model for parametric insurance of zerogeneration events at offshore wind farms, mainly focusing on the Norwegian shelf. The model employs a Hierarchical Bayesian approach to analyze the variability of these events, leveraging historical data. The research uses Markov Chain Monte Carlo simulations to estimate posterior Gumbel distributions for zero generation events on both monthly and regional scales. A significant aspect of the study involves using copulas to model co-dependency between wind farms within a portfolio, and employing Value-at-Risk and Expected Shortfall metrics for risk assessment. A crucial finding is the heightened risk in insuring portfolios of geographically proximate wind farms due to co-dependency, evident in increased premiums and risk metrics. Additionally, the thesis explores the impact of co-dependency on insurance premiums, noting a trend reversal in premiums based on a trigger event threshold. The research concludes that insurance companies can effectively utilize this pricing strategy for insuring multiple wind farm locations, encouraging offshore wind sector investment by securing operator revenue streams.