timeseries prediction for multiple departments

[Fierel]

HI,

I am trying to do a timeseries forecast prediction. my data set comprise of multiple variables which I separate to become univariate(in order to do the ARIMA prediction), it is also across multiple departments, (Please see example)

Is there a way to perform the calculations/forecast while iterating through the department and joining them back together.

Thank you very much for the help

[vsoni · Accepted solution]

The TS Forecast Factory tools should be able to help you here.

You can download them + see the documentation here: https://help.alteryx.com/20213/designer/ts-forecast-factory-tool

[Fierel]

This is actually amazing, although it highlighted a data problem. It solved the initial dilemma but now it raised a new one since it uses auto arima it seems to not identify seasonality which results in a straight line (constant) forecast, is there a workaround for this?

Thank you very much for the tip 🙂