Forecasting and Reconstructing Time Series

This workflow forecasts the monthly average sales in 2017 based on monthly average sales between 2014 and 2016 using dynamic deployment. The forecasting model is an ARIMA (0,1,4) model. The forecasted sales values consist of the forecasted residuals and restored seasonality and trend components.


This is a companion discussion topic for the original entry at https://kni.me/w/2ffBxw87KyFb4JbM