How to perform cyber risk assessment via cumulative logit models

Cyber risk assessment is a field with a growing amount of research. Information on monetary losses due to a cyber attack is difficult to gather due to the sensitive nature of the data. An evaluation of the severity of the attack on an ordinal scale is easier to obtain. We apply a cumulative logit model to identify the variables that mostly affect the severity of an attack. A systemic risk indicator based on social network analysis is proposed and included in the model.