Document Type : Research Paper
Authors
1 Associate Professor of Economics, Allameh Tabataba'i University
2 faculty of economics. allameh tabatab'i university
3 Master Student, Allameh Tabataba'i University
Abstract
The aim of the current research is to present a model for the risk analysis of 30 large companies in the Tehran Stock Exchange using the multivariate factor stochastic volatility model (MFSVM) in the framework of the non-linear state-space approach. In this framework, the volatility of stock returns is divided into two components, "volatility arising from latent factors" and "idiosyncratic risks", and the dynamic correlation matrix of the volatility of stock returns is estimated. In this regard, the weekly data of companies' stock returns during the period of Jan. 10, 2018 to Oct. 7, 2023 have been used. The results of the research indicate that, first- three hidden factors affect the volatility of stock returns. The first factor has affected stocks related to companies active in the oil products industry, chemical products, basic metals, mining, and investment funds. The second factor has had the largest impact on banks. The third factor has also had an effect on bank stocks to some extent. Second- the strongest posterior pairwise correlation exists between “GDIR” and “PTAP”, “PASN”, and “FOLD”, with correlation coefficients of 74%, 73%, and 71%, respectively. Additionally, “FOLD” exhibits a 69% correlation with both “PASN” and “PTAP”, as well as correlations of 66% each with “MSMI” and “MADN”. The weakest correlation coefficient is between “GDIR” and “BPAS” (-10%). Third BPAS (Pasargad Bank) experiences the lowest correlation with the stock network, while GDIR (Ghadir Investment Company) exhibits the highest correlation.
Keywords
- Idiosyncratic Risk
- Latent Factors
- Multivariate Factor Stochastic Volatility Model
- Dynamic Correlation
- Stock Market
Main Subjects
)