Data ready to be used

For econometric discussions not necessarily related to EViews.

Moderators: EViews Gareth, EViews Moderator

Sahara
Posts: 10
Joined: Tue Apr 17, 2012 11:32 am
Location: Hamburg

Data ready to be used

Postby Sahara » Thu Dec 06, 2012 6:18 am

I would like to know how to know that my data is good or ready to be used on a research. If I have an excel file with all the data and plot it on eviews, is it good to be used? Is there a way to be sure before continuing with my research?

Thank you for your time!

trubador
Did you use forum search?
Posts: 1520
Joined: Thu Nov 20, 2008 12:04 pm

Re: Data ready to be used

Postby trubador » Thu Dec 06, 2012 7:59 am

I think you are asking for preliminary data analysis, which depends on the nature and the purpose of your research. EViews has many diagnostic tools that can guide you throughout your research. However, you can never be 100% sure about the quality or feasibility of your data before digging into it.

Sahara
Posts: 10
Joined: Tue Apr 17, 2012 11:32 am
Location: Hamburg

Re: Data ready to be used

Postby Sahara » Thu Dec 06, 2012 8:26 am

I need to perform a johansen cointegration test. Therefore, I need to test for the unit root of series of Bonds. I have performed the ADF and all 6 series are considered to be non-stationary. However, when I perform at 1st difference, it seems to be stationary. My question is: I am able to use this data to perform the johansen cointegration test?
If the answer is indeed positive, do I have to save this results at 1st difference to use with the johansen test or can I just use select the johansen option and analyse the results? I hope I am being clear with my questions =/

Sahara
Posts: 10
Joined: Tue Apr 17, 2012 11:32 am
Location: Hamburg

Re: Data ready to be used

Postby Sahara » Thu Dec 06, 2012 8:47 am

In order to be more clear regarding my question about saving the 1st differencing series.
Can I do the command GENR "dGER = d(GER) ", " dES = s(ES) " and then use this new series to conduct my research on Johansen and Granger tests? Or do I use the previous series GER and ES? This might be a really silly question but it would help me to know the answer

trubador
Did you use forum search?
Posts: 1520
Joined: Thu Nov 20, 2008 12:04 pm

Re: Data ready to be used

Postby trubador » Fri Dec 07, 2012 4:16 am

You can use your data in levels, without the need to take differences.

Sahara
Posts: 10
Joined: Tue Apr 17, 2012 11:32 am
Location: Hamburg

Re: Data ready to be used

Postby Sahara » Fri Dec 07, 2012 9:33 am

Thank you for the tip!

Will the results for granger and johansen be different if I use the 1st difference level? If so, which one do you advice to use, the 1st difference or the levels?


Return to “Econometric Discussions”

Who is online

Users browsing this forum: No registered users and 2 guests