Dear all,
I have a problem on using Arima model, and really hope that you can help me to solve this.
This is the first time I use this kind of model, and sorry if I am asking something puerile.
I show you some graphs, which is named based on the parameters p,d,q of the model.
when I use dfuller test for the origninal variabel, in this case, is sxh_m, it gives me that the series is stationary. Hence, I don't use nonseasonal difference here. And when I read graphs of ACF and PACF, I choose AR(1) model. I don't know what I did is right or wrong?
Next, plotting graphs ACF and PACF of residuals of the AR(1). Personally, I think the series has a strong lag at lags 6,12 and then I should use Sarima. Am I right? And what should I do next?
Thank you all in advance!
I have a problem on using Arima model, and really hope that you can help me to solve this.
This is the first time I use this kind of model, and sorry if I am asking something puerile.
I show you some graphs, which is named based on the parameters p,d,q of the model.
when I use dfuller test for the origninal variabel, in this case, is sxh_m, it gives me that the series is stationary. Hence, I don't use nonseasonal difference here. And when I read graphs of ACF and PACF, I choose AR(1) model. I don't know what I did is right or wrong?
Next, plotting graphs ACF and PACF of residuals of the AR(1). Personally, I think the series has a strong lag at lags 6,12 and then I should use Sarima. Am I right? And what should I do next?
Thank you all in advance!
Comment