'Can anyone explain the working of statsmodels.tsa.arima.model.ARIMA() method?
I tried using statsmodels.tsa.arima.model.ARIMA method for Time Series forecasting in Python. The method has a parameter order=(p,d,q) that takes in the AR, difference and MA values. After doing the log transformation, I still need to get the cumulative differences (order=d).
Does the method itself compute the cumulative difference and map the predicted values back to original scale? What is the reason behind passing the d parameter in the method? How does it work?
If not, then what is the way to scale back the predicted values (after differencing d times) to the original space (that of the original Time Series)?
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|