Time Series Prediction

This workflow has the aim to build an auto-regressive model using the previous 24h*7 as seasonality template: - 24-hour seasonality template: the first week of the time series is used as a template for seasonality correction; - auto means usage of past of the same time series for prediction. No other external time series/data used; - Regressive refers to the mode used: either a linear or a polynomial regression.

This is a companion discussion topic for the original entry at https://kni.me/w/kyzNLnmNn0WWLeaY

I have been learning Knime for a month and i find it quite useful.
Can i ask in this case, why in the metanode “Prepare Data” , needs to use concatenate node so in a way its duplicated the data?

Hello @Xiaomeng and welcome to the forum.

I think you’re right, that the Concatenate node here (and the associated Column Filter node) are doubling up the data. You can remove them.

Thanks for bringing this to our attention!

(I’ll add that this workflow is a pretty old one, and I think it probably worked slightly differently when it was originally posted on the Example server. We’re making progress on a refresh of these older workflows, and I’ll add this one to the list.)

Thanks to the quick response!