Hello, Thanks a lot for the good feedback!

About the question, I do not understand if you are using the same dataset, but if you are using another dataset I completely understand you may find the numbers in (S)ARIMA different. Inside pm.auto_arima I used raw 'Production' column and set p=1 and D=1 because for this dataset differentiating once for the trend (p) and once for seasonality(D) was enough to make it stationary. Also I used max_order=6 to limit the sum of p+q+P+Q maximum to 6. If you leave it default then it is set to 5, if you set to None, then there will be no limit on it. I hope this answers your question and I recommend (if you haven't) to check also the source code of the article on my github. https://github.com/Idilismiguzel/Time-Series/blob/master/SARIMA-forecasting.ipynb