“…It can detect the noise and chaos including the structural breaks, regime-switching, and, most likely the nonlinear behavior of the data series easily (Hinton, 1992; Gupta and Kashyap, 2016; Alaloul et al , 2018; Marcos et al , 2020). Moreover, the results of the studies show that the stock market prophecies are quite achievable applying AI-driven hybrid nonlinear volatility models, and, it produces better results in comparison to the benchmark GARCH model and its counterparts (Gonzalez Miranda and Burgess, 1997; Roh, 2007; Atsalakis and Valavanis, 2009; Chen et al , 2010; Guresen et al , 2011; Lahmiri and Boukadoum, 2015; Gopal and Ramasamy, 2017; Chkili and Hamdi, 2021; Goel and Singh, 2021). Though, the models depicted in the present study are restricted to boundaries such as GARCH model is more effective in forecasting short-term volatility than long-term volatility; IGARCH, FIGARCH, HYGARCH model suitability in capturing long-memory volatility; EGARCH, TGARCH, GJR-GARCH model better account asymmetrical behavior in volatility; multivariate-GARCH model appropriateness in simultaneous computing volatility of multiple assets; high frequency based multivariate heterogeneous autoregressive model suitability for longer forecast horizons, etc.…”