@clyton91 wrote:
Hi, I would like some guidance with using ARIMA models (statsmodel library) in Python
I have a time series which is not stationary but can be considered stationary after taking the log10 transformation and 1st order differencing.
My question is, which series should I pass to ARIMA() when building the model?
ARIMA(ts)
ARIMA(ts_log)
ARIMA(ts_log_diff)Thank you!
I’m referring to this post - https://www.analyticsvidhya.com/blog/2016/02/time-series-forecasting-codes-python/
Posts: 1
Participants: 1
