Issues with seasonality and ARIMA terms.

#1
Good evening!

I am pretty new to the times series analysis and I am having a hard time analyzing on EVIEWS one where there is a strong seasonality component,.

1. Since the variance wasn't stationary I applied a log transformation to the series.

Initial statistics:
1521052413774.png

After the Log transformations:
1521052574644.png

Then I check if the series was stationary, and after checking the correlograms (below) I decided to run the X11 to deseasonalize the series.
1521052714319.png

Then I checked the correlogram to see if there was any evident pattern of seasonality.
1521052761028.png

The autocorrelation function didn't show any signs of seasonality, but a clear patter of non stationarity of the series, so I proceeded with the first derivation. Getting the following correlogram: (next post)
 

Attachments

#2
Good evening!

I am pretty new to the times series analysis and I am having a hard time analyzing on EVIEWS one where there is a strong seasonality component,.

1. Since the variance wasn't stationary I applied a log transformation to the series.

Initial statistics:
View attachment 129

After the Log transformations:
View attachment 130

Then I check if the series was stationary, and after checking the correlograms (below) I decided to run the X11 to deseasonalize the series.
View attachment 132

Then I checked the correlogram to see if there was any evident pattern of seasonality.
View attachment 133

The autocorrelation function didn't show any signs of seasonality, but a clear patter of non stationarity of the series, so I proceeded with the first derivation. Getting the following correlogram: (next post)
1521052911526.png

That looks to me there is an AR(2) component but when I run the test the value I get is not significant, so I am not too sure what to do from that point.
1521053163826.png

Anybody could help me?

Many thanks in advance!