Nonstationary Generalised Autoregressive Conditional Heteroskedasticity Modelling for Fitting Higher Order Moments of Financial Series within Moving Time Windows

Figure 8

as a function of the running parameter . Here, we show how we calculate the value of for a particular time window. The orange line is the value of for the empirical time window, , whilst the blue line shows , equation (23) when (equation (24)) and (equation (25)), for the GARCH-double-normal model. The intercept of the two lines shows the value of which allows us to model data for a certain median time and a certain time window within the GARCH-double-normal (1,1) model.