Investigating the social factors affecting shareholder retention based on modeling and predicting stock market fluctuations using heterogeneous autoregressive (HAR) models in Tehran Stock Exchange

In recent years, financial markets have faced many fluctuations and the uncertainty caused by these fluctuations has raised concerns among investors. Therefore, fluctuation modeling and its prediction in various research and practical financial issues, has been considered. (HAR) is listed on the Tehran Stock Exchange. In this research, fluctuation modeling has been performed using abundant data and with the help of HAR-RV family models, and the effect of adding the jump component on the volatility prediction efficiency of the entire Tehran Stock Exchange index has been investigated. The statistical population of the study includes companies listed in the Tehran Stock Exchange in the period 1391 to 1398. Statistical sample Using purposive sampling method and applying the existing restrictions, 162 companies were selected as the statistical sample. In terms of applied research orientation, the present study is a mixed research approach (qualitative / quantitative) and the purpose of exploratory research. Based on the results, it can be said that if HAR models were not the best predictive model, it was not the worst, and since based on three criteria, its performance was the best, so it can be said that it was a good and suitable model for forecasting; Another is that both modeling based on GARCH and its competitor model, ie GARCH, both performed better on the assumption of normal distribution than on the student distribution assumption for error sentences.