Approximation of the VaR by combination of HARQ model and extreme value theory in Tehran Stock Exchange

One of the important categories in investing, especially in the stock market, is to pay attention to the issue of risk management for investors and companies. One of the most widely used measures in risk calculation, which includes both time horizon and confidence level, is the value at risk (VaR), which has been highly regarded by researchers and risk analysts over the past two decades. There are various methods, including parametric, semi-parametric and non-parametric, for calculating the risk arc, which can be used depending on the type of data. In this paper, by combining the generalization of Heterogeneous autoregressive model (HAR), ie Heterogeneous autoregressive quarticity model (HARQ) and extreme value theory, we introduce an efficient method for calculating the risk value of some stocks present in the Tehran Stock Exchange and compare the advantages and comparison with other approaches such as parametric and non-parametric. We will address the disadvantages of the introduced method. The statistical population of this research is an example of active companies that were present in the stock exchange from 1392 to 1397. It is noteworthy that the half-hour time frame has been adopted in order to achieve higher accuracy. It is also a method of data analysis using MATLAB software.