Asked about Forecasting models, from simple to learned

Why evaluate at multiple forecast horizons?

An answer that works

Error grows with the horizon, and an averaged single number hides how quickly accuracy decays. Evaluate at each horizon you will actually use.

Reading an answer is not being able to give one. The section that teaches this is Forecasting models, from simple to learned, 4 hours of Time series.

Learn this properly

Also asked about Time series

Why can you not use ordinary cross-validation on a time series?What is the first thing you do on a forecasting problem?How do you build a rolling average feature safely?Would you use a neural network or a classical model for forecasting?How would you evaluate a demand forecasting model?You need forecasts for 50,000 products, many with little history. How do you approach it?