“…The most widely used method is the standard MLP [19,20,32,41,42,51,91,92,94,96,97], followed by the adaptive network-based fuzzy inference system (ANFIS) [19,90,95], radial basis function network [20,24,100], and autoregressive models like ARMA or ARIMA [20,22,24,90]. Other models include LSTM [17], linear regression [50], extreme learning machine [22,50], CNN [50], Bayesian neural network [26,99], exponential GARCH [90], echo state neural network [27], Elman neural networks [18], and support vector regressors [20]. It is important to note that in many of the approaches, the hybrid method does not consider a single forecasting model but combines several of them [19,20,24,50,90,97].…”