Modeling Volatility of Russian Financial Market in Periods of Crisis

Today's economy is a developing field where everything changes – prices are no exception. Such conditions are associated with risks and one of the most common measures of risk is volatility. This paper analyses different models of volatility for the RTS index return during the financial stability and during the crisis. Models that are considered in the diploma include popular GARCH-type models and high-frequency data models. Overall, 18 models were described and analyzed: GARCH, EGARCH, TGARCH and HAR-RV, HAR-RR, HAR-RV-J, HAR-RV-CJ, HAR-RV-L and their interpretation with the logarithms and square root. On the basis such of indicators as RMSE, MAE and MAPE, it was concluded that high-frequency data models are better. In addition, in the pre-crisis period the best model includes asymmetry. That is, different returns (positive and negative) have different impact on volatility. However, during the crisis, the inclusion of jumps in the model positively affects its performance.