Dear all,
I have a historical gold prices time series from 22 September 2008 to 17 September 2021. I would like to predict the gold prices for the ten next days using machine learning methods.
I calculated the autocorrelation function for determining the inputs of the predictive models:
m = ACF[data, 800];
acfS = ListLinePlot[m, ImageSize -> 500,
PlotStyle -> {Orange, Thickness[0.005]},
AxesStyle -> Directive[Black, 12], PlotRange -> All]
It is clear that using the antecedent 400 values as input for the predictive model does not make sense. What can I do for solving this problem? Attached are the data.
I appreciate your time.
Alex
Attachments:
