Hi guys,
I need your assistance, I am a newbie for Eviews :)
There is a model that I must estimate but I don't know how.
Here is the model.
log (tradeij) = β0 + β1log(GDPi) + β2log(GDPj) + β3log(distanceij) + β4RTAij + εij
trade ij = trade between country i and country j (e.g. between Turkey and Germany)
β0, β1, β2, β3, β4 = unknown parameters
log(GDPi) = Gross domestic product for country i
log(GDPj) = Gross domestic product for country j
log(distanceij) = distance between country i and country j
RTAij = Regional Trade Agreements between country i and country j ( RTA takes only 0 or 1 values)
εij = error term
Here is the data examples between Turkiye and Germany
obs trade GDP_Germany GDP_Turkiye Distance RTA
1990 6.560.405 1.547.026 149.197 2.036.075 0
1991 6.644.949 1.815.061 147.756 2.036.075 0
1992 7.414.893 2.066.729 156.173 2.036.075 0
1993 8.187.120 2.005.557 177.004 2.036.075 0
1994 7.579.902 2.151.025 128.089 2.036.075 0
1995 1.058.376 2.524.949 166.443 2.036.075 0
1996 1.300.009 2.439.346 178.061 2.036.075 1
1997 1.327.465 2.163.233 186.061 2.036.075 1
1998 1.277.562 2.187.484 197.587 2.036.075 1
1999 1.135.464 2.146.432 181.690 2.036.075 1
2000 1.237.805 1.905.795 198.230 2.036.075 1
2001 1.070.239 1.892.595 143.096 2.036.075 1
2002 1.291.034 2.024.060 182.973 2.036.075 1
2003 1.693.790 2.444.284 240.596 2.036.075 1
2004 2.126.094 2.744.221 302.561 2.036.075 1
2005 2.308.894 2.791.737 362.461 2.036.075 1
2006 2.431.924 2.897.032 392.424 2.036.075 1
2007 2.953.319 3.080.553 432.512 2.036.075 1
2008 3.163.895 3.184.131 455.538 2.036.075 1
P.S.: You can find the data examples in the attached file.
I use this equation but I get this error.
Equation = log(trade) c log(GDP_Germany) log(GDP_Turkiye) log(Distance) rta
Error = Near singular matrix
When I remove the "log(Distance)" , there isn't any error.
How can I insert "log(Distance)" in to the equation.
Thanks for your support.
B.R.
Gravity Model
Moderators: EViews Gareth, EViews Moderator
-
thehighyam
- Posts: 4
- Joined: Sat Dec 12, 2009 2:14 pm
Gravity Model
- Attachments
-
- Data.xls
- (17.5 KiB) Downloaded 834 times
-
startz
- Non-normality and collinearity are NOT problems!
- Posts: 3796
- Joined: Wed Sep 17, 2008 2:25 pm
Re: Gravity Model
You can't.
Distance is a constant in your data, so there's no way to find out what effect it has on trade.
Distance is a constant in your data, so there's no way to find out what effect it has on trade.
-
thehighyam
- Posts: 4
- Joined: Sat Dec 12, 2009 2:14 pm
Re: Gravity Model
Hi
Thank you for your quick reply startz
Well, is there any other way to include this constant value? Cos in other articles, they inserted this constant value in their gravity model. I can't understand how but they could.
Thank you again.
Thank you for your quick reply startz
Well, is there any other way to include this constant value? Cos in other articles, they inserted this constant value in their gravity model. I can't understand how but they could.
Thank you again.
-
startz
- Non-normality and collinearity are NOT problems!
- Posts: 3796
- Joined: Wed Sep 17, 2008 2:25 pm
Re: Gravity Model
You have to use more than one pair of countries, so that distance isn't the same for all observations.
Re: Gravity Model
Hi startz,
I wonder, how do I follow your suggestion in eviews - to use more than one pair of countries? Can you give the command to follow? (I'm having the same problem with thehighyam).
I wonder, how do I follow your suggestion in eviews - to use more than one pair of countries? Can you give the command to follow? (I'm having the same problem with thehighyam).
Re: Gravity Model
Owh, thank you so much for the help in advance
-
startz
- Non-normality and collinearity are NOT problems!
- Posts: 3796
- Joined: Wed Sep 17, 2008 2:25 pm
Re: Gravity Model
Unless you exclude them, EViews will use all the country pairs you have. How many countries are in your data?
Re: Gravity Model
it seems to me that you are trying to estimate potential effects of preferential agreements using the gravity model.
in that case, you have to use panel data which incorporates not only cross-sectional analysis but also time dimension as well.
it means you need to add other country-pairs into your data for comparison purpose.
country i and j can be turkey and germany but add other pairs like turkey-greece, turkey-italy, turkey-cyprus into your model and then distance variable will not be constant any more. it also allows other dummy variables such as common border and language to be included in the model.
distance will turn out to be very significant so i'd recommned you not to omit it from the regression.
in that case, you have to use panel data which incorporates not only cross-sectional analysis but also time dimension as well.
it means you need to add other country-pairs into your data for comparison purpose.
country i and j can be turkey and germany but add other pairs like turkey-greece, turkey-italy, turkey-cyprus into your model and then distance variable will not be constant any more. it also allows other dummy variables such as common border and language to be included in the model.
distance will turn out to be very significant so i'd recommned you not to omit it from the regression.
-
estrella bernabeu
- Posts: 16
- Joined: Thu Mar 25, 2010 8:00 am
Re: Gravity Model
I'm opering with the gravity model, and my professor told to do the hausman test and to do it i have to proceed by the OLS test and to know if the model is random or fixed and the run the hausman test.
I don't understand exactly what the prof is syaing, so i need your help in case to pass this step.
I don't understand exactly what the prof is syaing, so i need your help in case to pass this step.
-
trieandini
- Posts: 1
- Joined: Mon Jun 14, 2010 7:40 am
Re: Gravity Model
hi, :D it really nice i found this link...
please help me...
i also use gravity model with case indonesian bilateral trade with ASEAN members countries...i dont know how to use eviews to regress my data...
the model similar with thehighyam, but i dont know how to get the result.. do you have any suggestion for me???may be the books?? or anything else???
thank you for your attention..please help me...
please help me...
i also use gravity model with case indonesian bilateral trade with ASEAN members countries...i dont know how to use eviews to regress my data...
the model similar with thehighyam, but i dont know how to get the result.. do you have any suggestion for me???may be the books?? or anything else???
thank you for your attention..please help me...
Who is online
Users browsing this forum: No registered users and 2 guests
