Gravity Model of international Trade

For technical questions regarding estimation of single equations, systems, VARs, Factor analysis and State Space Models in EViews. General econometric questions and advice should go in the Econometric Discussions forum.

Moderators: EViews Gareth, EViews Moderator

GiMa
Posts: 6
Joined: Wed May 22, 2019 12:07 am

Gravity Model of international Trade

Postby GiMa » Tue May 28, 2019 10:27 am

Hey there!
For my thesis I have to estimate a Gravity Model of international Trade. I use over 30 countries during the time of 1990-2017. I have variables which vary over time (GDP and Exports) and constant variables (distance) as well as constant and changing dummies. I transferred my Data in 3d panel form to EViews, but now I am lost how to estimate the model.
I tried: equation eq.ln @log(exports_ij) c @log(GDP_ij) @log(distance) @dummy
but it is not working.

Further my GDP data is in current USD and my Exports in thousand current USD, how can I adjust those values?


Thank you in advance!

startz
Non-normality and collinearity are NOT problems!
Posts: 3499
Joined: Wed Sep 17, 2008 2:25 pm

Re: Gravity Model of international Trade

Postby startz » Tue May 28, 2019 10:34 am

I don't think @dummy is a function. Perhaps you were looking for @expand?

GiMa
Posts: 6
Joined: Wed May 22, 2019 12:07 am

Re: Gravity Model of international Trade

Postby GiMa » Tue May 28, 2019 10:39 am

Thank you!

So it is: @log(exports_ij) c @log(GDP_j) @log(distance) @expand(dummy)?

startz
Non-normality and collinearity are NOT problems!
Posts: 3499
Joined: Wed Sep 17, 2008 2:25 pm

Re: Gravity Model of international Trade

Postby startz » Tue May 28, 2019 10:42 am

Probably best to read the help documentation.

GiMa
Posts: 6
Joined: Wed May 22, 2019 12:07 am

Re: Gravity Model of international Trade

Postby GiMa » Wed May 29, 2019 9:17 am

I did and I figured out an equation that is leading me to an output:
log(exports_ij) c log(gdp_i) log(gdp_j) log(distw) log(dist) log(distwces) log(distcap) combo cefta_new cefta_92 lang_off colony curcol col45 smctry

but somehow my cefta_new coefficient is negative (cefta_92 is positive) and my distance coefficient positive. I can't find a mistake in my data.

startz
Non-normality and collinearity are NOT problems!
Posts: 3499
Joined: Wed Sep 17, 2008 2:25 pm

Re: Gravity Model of international Trade

Postby startz » Wed May 29, 2019 10:00 am

It doesn't look like you've included any dummies.

GiMa
Posts: 6
Joined: Wed May 22, 2019 12:07 am

Re: Gravity Model of international Trade

Postby GiMa » Wed May 29, 2019 10:06 am

combo cefta_new cefta_92 lang_off colony curcol col45 smctry are the dummies

startz
Non-normality and collinearity are NOT problems!
Posts: 3499
Joined: Wed Sep 17, 2008 2:25 pm

Re: Gravity Model of international Trade

Postby startz » Wed May 29, 2019 11:20 am

I don't know why you are getting that result. Perhaps someone who has estimated gravity models will have an insight. But you'll have to give clear definitions of all your variables.

GiMa
Posts: 6
Joined: Wed May 22, 2019 12:07 am

Re: Gravity Model of international Trade

Postby GiMa » Thu May 30, 2019 12:25 pm

I checked my variables again and I am really desperate now. I'll attach my excell data.

I would be more than thankful if anyone could help me!

startz
Non-normality and collinearity are NOT problems!
Posts: 3499
Joined: Wed Sep 17, 2008 2:25 pm

Re: Gravity Model of international Trade

Postby startz » Thu May 30, 2019 12:34 pm

No one knows what your variable names mean.


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 10 guests