Spurious regressions?

I refer to the fpp2 testbook: 9.3 Forecasting | Forecasting: Principles and Practice (2nd ed)
In this example, the electricity demand and the temperature series are both non-stationary. Elsewhere in the book, and as we know commonly, the advise is to either check for co-integration, or to use differencing to make the variables stationary.

Hence, isn't doing the regression a problem, as it is essentially based on spurious regression? How do I justify this regression?

Referred here by Forecasting: Principles and Practice, by Rob J Hyndman and George Athanasopoulos

the electricity demand and temperature example you point to are addressed with differencing though, as that is a component of ARIMA.

Thanks a lot for pointing that out: I missed that. My bad!

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.