4 Week Moving Average Forecasting Model

Hello all,

I am fairly new to Knime and am trying to create a 4 week moving average forecasting model. Here is the workflow I have so far (the two calculated fields are YoY and 4 Week MA of the YoY)

I want to somehow use the recursive loop node (and I believe also the polynomial regression learner+predictor) to forecast X periods ahead (lets say 13 weeks into the future) of my most recent week of actuals. Can anyone suggest how I should go about doing this? Any help would be much appreciated.

Thank you!

Hi @lt476

I just deleted the uploaded workflow because it contains username and password to your database.

Could you make me an example with a small artificial data set?

1 Like

Hi @Iris,

My apologies and thank you for doing so.

Here are screenshots of the workflow and some of the calculated fields I mentioned.




Final data table:

Hello @lt476,

have you made some progress with your model? You can check following example on KNIME Hub which features recursive loop and regression learner to do some forecasting on time series data set.

Br,
Ivan

1 Like

This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.