Advent of Code is back! Unwrap daily challenges to sharpen your Alteryx skills and earn badges along the way! Learn more now.

Alteryx Designer Desktop Discussions

Find answers, ask questions, and share expertise about Alteryx Designer Desktop and Intelligence Suite.

Measures to compare Arima models to already existing predictions

AnRo
6 - Meteoroid

Hello!

 

Every year our controlling deparment needs to make a forecast on monthly sales volume for the upcomming year and i want to help them with an Arima model.

 

My question isn't about the Arima model(s) itself though, rather on how to best compare them to the manual work done by my colleagues in the past.

 

I want to have a baseline to see how the models i can come up with compare to the human made forecast. While comparing two or more Arima models (on the same data) is relatively easy with the measures put out by Alteryx, i haven't found a clear way to compare the models to "already existing" forecast data.

 

I thought about calculating MAD, RMSE and SMAPE to compare the Arima results to the human made ones but i'm not sure if those will give me an robust understanding on if my models are better/worse (+and by how much) than the forecast created by my colleagues.

 

I hope i was able to convey what i am interested in understanding.

 

Thank you 

 

AnRo

0 REPLIES 0
Labels