Temporal series. Generation of Models Automatically.

KNIME Analytics Platform
1

Hello it can be done in KNIME, the following: enter several series (such as data related to product ranges, regions or warehouses), and generate a separate model for each target series. For example, if income from the eastern region is found to fit an ARIMA model, but the western region fits only a simple moving average, each region will be scored using the appropriate model.

thanks to everyone

Marcos

2

Hi there @macosta,

you would like to have workflow that runs data through several models and highest scoring model to be chosen?

Think this can help you get started with it in KNIME:
https://www.knime.com/blog/the-knime-model-process-factory

Br,
Ivan

1 Like
3

Hi Ivan, if that would like to have a workflow that runs the data across multiple models and the highest scoring model for. With this flow can I do it 50_Applications / 26_Model_Process_Management?

4

Hi @macosta,

yes. But you will have to modify it to your needs.

Also you can check this workflow which does same:

Br,
Ivan

2 Likes
5

Hello Ivan, I try it and I will tell you if it is what I was looking for. Thank you!!

1 Like
6

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