I would like to share a new article on some of the methods and pitfalls of time series forecasting: “A Time Series Apologia”. In it I work the seemingly simple problem of forecasting a noisy copy of
sin(t). The purpose of the article is to demonstrate using ARIMA methods, and to show that it is okay to also try non-ARIMA methods. We also share a very important, but under-taught, method for choosing the MA degree of an ARIMA model.
NOAA Tide Predicting Machine No. 2