Suppose I have eleven times series X1, X2, X3....X11. All of them are stationary after the first normal differentiation and first seasonal diff. I would like to group the data in just three regressors for making an ARIMAX model. Suppose new drivers are Z1=x1+x2+x3+x6 Z2=X4+X5+X7 and Z3=x6+x8+x9+x10+x11.
The question is what is more convenient and correct to do: To make the groups before making the differentiation or grouping the data once all Xi are stationaryzed?
Thanks
Referred here by Forecasting: Principles and Practice, by Rob J Hyndman and George Athanasopoulos