/sci/ - Science & Math

Why exactly is it that a non-stationary time series can be made stationary by differencing the series?

Is there any significance in the solution to the difference equation for forecasting purposes?

I'm constructing a time series forecast for inflation the US

Time series methods are underrated imo, much more fun than non-regression stats

Will we ever be able to truly tell the future /sci/?

Will we ever at least be able to guess with a reasonable degree of accuracy?

Is there any real way to tell if a process is a moving average, AR or ARIMA process just by looking at the graph or is it always best to go by Akaike's information criteria?

also /stats/ general