Hello everyone
i’m new to knime and i’m trying to predict sales figures for the next year, based on the last 10 years, how can i get a better R value ? Does anyone have any idea what i am doing wrong?
I have searched the forum for a suitable solution but unfortunately have not found one.
Thank you very much for your Help
best regards tom
sales_data10Years_updated.xlsx (11.8 KB)
Sarima Model 1.0 (Community Hub).knwf (135.5 KB)
Hi @TomM2025 -
Sorry for the delayed reply here. Were you able to make any additional progress?
I’m not a time series expert by any means, but one thing I noticed is that you are not applying the flow variables created by the Loop Start node to the Learner - so what happens is you just run a single series of variables several times.
By applying the flow variables properly, you can improve the R^2 value to 0.64. I’m not sure what your target is there but it’s a bit better than before.
