[Python] Determining ARIMA Model Parameters for Daily Time Series Forecasting

I am currently working on time series forecasting at the daily level using the ARIMA model.

As part of this process, I need to determine the appropriate values for the parameters p,d and q.

I have couple of questions regarding this:

1.Given that we are working with daily forecasting,I suspect that the values of p,d and q may be larger. Could you please advise me on the recommended range for these values?

1. In terms of identifying the values of p,d and q, I am unsure whether to rely on ACF or PACF graphs or to use Auto ARIMA approach. Which method would you recommend as the better option?

Continue reading...