Cyber Insurance Rate Making Based on Markov Model for Regular Networks Topology

Abstract

Abstract Computers and Internet play a key role in the processes of data transaction, data exchange and data storage in a business. Malicious software or computer viruses is one of the biggest threats that can attack computer networks and cause huge losses due to loss of data and information. As a way to transfer risk, cyber insurance requires precise and appropriate calculations even though many challenges are faced including the effects of differences in network structure. Standards of cyber insurance that have not been established as in the mortality table for life insurance open the possibility to set a standard calculation based on network structure by determining cyber insurance rates. This study uses a general susceptible-infectious-susceptible model with Markovian property to simulate the process of spreading computer virus and calculate the total loss for each computer. Rate making will consider the number of infected neighbours on a node as an exposure to set insurance rates on regular network topology. The simulation process shows that the rates at each node are affected by the probability of initial infection, the degree of each node on the network, and the parameters of infection or recovery.