Graphical modeling has mainly been limited to distributions that lead to severe underestimation of large risks and, therefore, to unsuitable models in the context of risk assessment. This thesis introduces a recursive max-linear structural equation model that finds its application in situations where extreme risks may propagate through a network, for example, when modeling water levels in a river network. We address structural properties, statistical issues, identifiability questions, and algorithms to apply the new model to data in the future.
«
Graphical modeling has mainly been limited to distributions that lead to severe underestimation of large risks and, therefore, to unsuitable models in the context of risk assessment. This thesis introduces a recursive max-linear structural equation model that finds its application in situations where extreme risks may propagate through a network, for example, when modeling water levels in a river network. We address structural properties, statistical issues, identifiability questions, and algori...
»