## test for seasonality

For questions regarding the import, export and manipulation of data in EViews, including graphing and basic statistics.

Moderators: EViews Gareth, EViews Jason, EViews Steve, EViews Moderator

christine

### test for seasonality

Dear all,

I've just created my account on this forum. It is extremely interesting.
I'm working on my final paper and i have a question. I hope you can help me regarding seasonality. How can I test whether my time series are seasonal or not? I know that I should use TramoSeats, ARIMA, etc for seasonal adjustments, but i wanna know if there is any method to test for seasonality (WITH THE EXCEPTION of the graphic method), not for adjustments. For example, taking the case of Consumer price index, looking at the graphic I cannot tell whether it is seasonal or not. But when I compute TramoSeats for instance, it works. I mean Eviews does adjust my series.

Thanks,

Best wishes,

Christine

startz
Non-normality and collinearity are NOT problems!
Posts: 3400
Joined: Wed Sep 17, 2008 2:25 pm

### Re: test for seasonality

christine wrote:Dear all,

I've just created my account on this forum. It is extremely interesting.
I'm working on my final paper and i have a question. I hope you can help me regarding seasonality. How can I test whether my time series are seasonal or not? I know that I should use TramoSeats, ARIMA, etc for seasonal adjustments, but i wanna know if there is any method to test for seasonality (WITH THE EXCEPTION of the graphic method), not for adjustments. For example, taking the case of Consumer price index, looking at the graphic I cannot tell whether it is seasonal or not. But when I compute TramoSeats for instance, it works. I mean Eviews does adjust my series.

Thanks,

Best wishes,

Christine

One method is to regress the series on dummies for each season and ask if they are significantly different.

Did you use forum search?
Posts: 1518
Joined: Thu Nov 20, 2008 12:04 pm

### Re: test for seasonality

Seasonality is a complicated subject and there are numerous different tests are available for the detection. As Startz said, the simplest way is to regress your dependent variable against seasonal dummies. For monthly data you can do the following in your equation object:

Code: Select all

`y @seas(1) @seas(2) @seas(3) @seas(4) @seas(5) @seas(6) @seas(7) @seas(8) @seas(9) @seas(10) @seas(11) @seas(12)`

Significant coefficient values will mean that you have deterministic and stable seasonality in those periods.

Another way is to check the report generated by Census X12 procedure when you run the seasonal adjustment. In that rich report, you will find the results of the tests for both stable and moving seasonality. You can also find test results for the presence of residual seasonality.

flourence
Posts: 5
Joined: Sat Oct 21, 2017 3:56 am

### Re: test for seasonality

what if I dont have ind and dep. variable . only my data . can i use the code :
y = own_use @seas(1) @seas(2) @seas(3) @seas(4) @seas(5) @seas(6) @seas(7) @seas(8) @seas(9) @seas(10) @seas(11) @seas(12) where own use is my data and y is the name of my file ?? ty

flourence
Posts: 5
Joined: Sat Oct 21, 2017 3:56 am

### Re: test for seasonality

After browsing in how to detect seasonality in data i found out that if your AC in correlogram shows up and down movement repeatedly it shows seasonality . Example is this. Can someone verify this
Attachments
AC.png (333.63 KiB) Viewed 2516 times

### Who is online

Users browsing this forum: No registered users and 5 guests