Forecasting and Reconstructing Time Series – KNIME Community Hub

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

Dear KNIME Community, i have some questions about forecast. I have some non linear development and i want to make a forecast for this sales.

So i think that ARIMA/SARIMA Model would not fit here. Can you please recommend some other methodes oder models that i can use in this case?

Thanks a lot!