Using Confidence Intervals for finding confidence of forecasting model

Question

My question is quite a general one.

Can I use the width of the confidence interval (let's say a 95% confidence interval) to find out how confident my model is while doing time series forecasting? I am thinking that as we go on forecasting in the future, the width of the confidence interval increases, and maybe after a threshold we can say that model is not confident enough to do forecasting. Is this a correct line of thinking?

Otherwise is there any other way I can use to determine how confident my model is and hence how much further in the future I can forecast?

Answer 1

The basic question is correct - as a forecast goes farther out in time, the confidence intervals become wider - sometimes extremely fast - thus reducing the confidence in the conclusions provided by the model(s).

The book Forecasting Principles and Practice has a section on this exact point. That section is: Least Squares Estimation

The relevant sections state that t-values and p-values are of "limited interest". Instead, the book covers several methods to evaluate goodness of fit for time series models (RMSE, AIC, AICc, etc.) The authors also created R packages that are extremely good for doing time series work, and I recommend the fpp3 package very highly. The book is available online from the authors at:

https://otexts.com/fpp3/