I have a time series of data on the flow of a river per month from January 926 to December 2022. When I apply the Sarima model to this data, I get 100% agreement with the values I have, and when I do modelling with linear regression, random forest and nearest neighbour models, it tells me the difference between the received and given data. .
Where am I going wrong? Am I using the Sarima model correctly?
