Financial market participants are constantly exposed to uncertainty and investment risk. Predicting and calculating risk is one of the most important issues in the field of financial issues. Reviewing the financial crises of recent years, it can be inferred that one of More
Financial market participants are constantly exposed to uncertainty and investment risk. Predicting and calculating risk is one of the most important issues in the field of financial issues. Reviewing the financial crises of recent years, it can be inferred that one of the reasons for these crises is the excessive attention to the repetitive central data and the lack of attention to the extreme data. In other words, in the analysis of financial data, the end part of the distribution should also be considered. The purpose of this study is to provide a model for tail risk estimation using extreme value mixture models. Accordingly, four one-tailed models and one two-tailed model in two simple functions and GARCH have been used. Modeling is based on three categories of data. The studied data include total index, price index (homogeneous) and index of top 50 companies. According to the obtained results, simulation of models with GARCH significantly improves the performance of models and reduces the error rate of simulated data in GARCH-based models. The findings also indicate that two-tailed models are more accurate than one-tailed models.
Manuscript profile
