On computing probabilities of dismissal of 10b-5 securities class-action cases

详细信息    查看全文

• 作者：Sumanta Singha ; ^{sumanta.singha@ku.edu}Author Vitae ; Steve Hillmer ^{hillmer@ku.edu}Author Vitae ; Prakash P. Shenoy ^{pshenoy@ku.edu}Author Vitae
• 关键词：Probability ; Logistic regression ; Naï ; ve Bayes ; Hybrid model ; 10b-5 securities class-action cases
• 刊名：Decision Support Systems
• 出版年：2017
• 出版时间：February 2017
• 年：2017
• 卷：94
• 期：Complete
• 页码：29-41
• 全文大小：1842 K
• 卷排序：94

文摘

A model for computing probability of dismissal of class-action cases is proposed. The prediction model is a hybrid model of naïve Bayes and logistic regression. Hybrid model predicts better than the naïve Bayes and logistic regression model. Features that are significant for dismissal are also identified. This model is useful for insurance underwriters, risk managers, policy-makers and researchers.