Ai Algorithms for Fitting GARCH Parameters to Empirical Financial Data

We use Deep Neural Networks to reconstruct GARCH parameters for empirical financial time series. The algorithm we develop, allows us to fit autocovariance of squared returns of financial data, with certain time lags, the second order statistical moment and the fourth order standardised moment. Our results show that the fitting of a GARCH model is only valid for time series of a certain duration. Analysing different sample sizes, we observe that we cannot fit GARCH parameters to time windows shorter than a minimum duration