MODELING THE BIVARIATE DEPENDENCE AND DETERMINING THE INSURANCE PREMIUM OF CYBER INSURANCE BASED ON CYBERATTACK DATA

• Main Author: Jonathan Haryono, Marcell
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/76273
• Institution: Institut Teknologi Bandung

It was assumed that different risks of cyberattacks are correlated with each other; however, research on the correlations between the risks of cyberattacks is still limited. In this final project, modeling the bivariate dependence between cyberattack risks on cyberattack data in the United States is carried out. The cyberattack data was obtained from the Privacy Rights Clearinghouse (PRC), which contains the number of breaches in the United States for the period of January 2005 to October 2019. The data was analyzed cross-sectionally, namely: between types of cyberattacks for cross-industry; and between industry types for cross-breach. Copula is used to model the bivariate dependencies between types of cyberattacks and between types of industries. After the bivariate dependency model is obtained, a model for the total number of breaches is constructed using the Loss Distribution Approach (LDA). The total number of breaches model is converted into an aggregate financial loss model which is used to calculate the cyberattack insurance risk premium. The Cross-industry and cross-breach risk premiums are determined using the Conditional Value-at-Risk risk measure.