Accelerating the local outlier factor algorithm on a GPU for intrusion detection systems
Proceedings of the 3rd Workshop on General-Purpose Computation on Graphics Processing Units
| Hi-index | 0.00 |
The financial operations of Taiwanese companies are becoming increasingly complex as a greater number of products are introduced into the market as a result of financial deregulation and recent reforms in Taiwan's financial markets. As financial statements are not able to fully reflect the actual state of companies' finances, many crises have occurred. Many investors are suffering from these crises, due to framing effect. This study investigates the outlying behavior of financial activities by using the outlier mining method to build models to predict financial crises. It uses local outlier factor (LOF) values to measure the outlying behavior among peer groups to gauge the financial performance of companies. This study tests its model on Taiwan's publicly-listed IC manufacturers and CDR makers. The result of the empirical study showed that the LOF value indicated the financial anomalies of these companies and confirm that this model can effectively provide advance warning to investors. Key Words: Outlier Mining, Local Outlier Factor, Financial Anomaly, Framing Effect