Document Type : Original Research Paper

Authors

• S. Asadi ¹
• A. Albadvi ²
• A. Husseinzadeh Kashan ³

¹ Department of Financial Engineering, Tarbiat Modares University, Tehran, Iran

² Department of Marketing and Electronic Commerce, Faculty of Industrial and Systems Engineering, Tarbiat Modares University, Tehran, Iran

³ Department of Economic and Social Systems, School of Industrial and Systems Engineering, Tarbiat Modares University, Tehran, Iran

Abstract

The challenge that the insurance companies' financial system is facing today is to understand the concept of risk and then measure and quantify the risk. One of the important risks of an insurance company is the market risk caused by investment. The main purpose of this article is to eliminate the shortcomings and defects of the regulations on how to calculate and monitor the financial solvency of insurance institutions and to take into account more precisely the characteristics of financial time series to estimate the value at risk of the investment portfolio (shares of stock exchange companies, foreign exchange accounts, and real estate). First, we use GARCH models to model the marginal distributions of the time series of the logarithm of returns. Then, using the genetic algorithm meta-heuristic method, we model the distribution sequences to obtain the best threshold in the Frein value theory and use the detailed function to model the correlation between the marginal distributions. The post-testing methods show that the proposed model performs better than the traditional model of historical simulation and the results obtained from the T-Student detailed function are more acceptable and the market risk coefficient was equal to %403.9.

Keywords

• Solvency
• Market Risk
• value at risk
• GARCH models
• Extreme Value Theory (EVT)
• Copula functions
• Genetic algorithm