报告时间：6月25日上午10：00-11：00
报告地点：行健楼526
Abstract: For cyber risk management, a cluster-based method is developed to investigate the risk of cyber-attacks in the continental United States. The proposed analysis considers geographical information on cyber incidents for clustering. By clustering state-based observations, the frequency and severity of cyber losses demonstrate a simplified structure: independent structure between inter-arrival time and size of cyber breaches. The independence between frequency and severity is significant at the state level instead of the national level. It is shown that the cluster-based models have a better fitting and are more robust than the aggregate model, where all incidents are considered together. To detect fraud insurance claims, we propose a new variable importance methodology incorporated with two prominent unsupervised deep learning models, namely, the autoencoder and the variational autoencoder. Each model's dynamics are discussed to inform the reader on how models can be adapted for fraud detection and how results can be perceived appropriately with a greater emphasis placed on qualitative evaluation.
金卓，博士，澳大利亚麦考瑞大学精算与商业分析系教授，精算和商业分析系的研究主任，北美准精算师（ASA）。历任澳大利亚墨尔本大学经济系精算中心讲师，高级讲师，副教授。研究方向为随机最优控制，随机系统的数值方法，精算学，数理金融，机器学习。在国际期刊发表70余篇论文，期刊包括Insurance Mathematics and Economics, European Journal of Operational Research, Journal of Risk and Insurance, SIAM Journal on Control and Optimization, Automatica, ASTIN: Bulletin and Scandinavian Actuarial Journal.
