Problems with SARIMA and ARIMA

KNIME Analytics Platform
1

Hi everybody,
I have three problems after the new ARIMA and SARIMA model updates:

  1. I can’t use the AUTO-SARIMA node correctly and I don’t understand if it requires a detrended and seasonally adjusted time series (“Residual”) or the original time series (“Signal”)

  2. In the past, the best parameters could be displayed in the “Port 2” output of the AUTO-ARIMA LEARNER. Now how can I view them?

  3. There are many different workflows, which use different systems to restore trend and seasonality. Which is the most correct?

Could anyone help me? @Corey I have used part of your templates, I hope I understood and used them correctly.

Problem with SARIMA and ARIMA.knwf (3.8 MB)

2

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.