Hello everyone ,
I am estimating gravity model with 3-dimensional panel data set (exporter, importer and year), including 24 countries and 18 years (2000-2017), so there are 24x23x18=9936 observations. I have created some fixed effects as follow:
For time fixed effect: tab year, gen(time_fixed)
For exporter fixed effect: tab Exporter, gen(exp_fixed)
For importer fixed effect: tab Importer, gen(imp_fixed)
For pair fixed effect:
egen pair_id = group(Exporter Importer)
tab pair_id, gen(pair_id_fixed)
My questions are:
1. if I want to run pooled OLS, should I use
reg Y X1 X2 X3
or
tsset pair_id year
xtreg Y X1 X2 X3
2. if I want to add only time fixed effect, should I use
reg Y X1 X2 X3 time_fixed*
or
tsset pair_id year
xtreg Y X1 X2 X3 time_fixed*
3. if I want to include time fixed effect, exporter fixed effect, importer fixed effect and pair fixed effect, should I use
reg Y X1 X2 X3 time_fixed* exp_fixed* imp_fixed* pair_id_fixed*
or
tsset pair_id year
xtreg Y X1 X2 X3 time_fixed* exp_fixed* imp_fixed*, fe
Thank you very much for your kind attention
Related Posts with Gravity model estimation
Hausman testI would like to ask you for an advice about the hausman test. The results show that there is now sta…
when importing excel into stataHello, I'm trying to import excel data into stata and it seems to misplace some values. Most values…
Comparing coefficients across two models (same X data, but slightly different Ys)Hi Stata Forum, I am checking if there is a simple way to compare two coefficients from xtreg. Take…
How to Deal with Outlying DataHi All, I'm working with a data set that measures the black-white disparity in proportion of popula…
Filling in years by groupHi, I have a dataset of the following form: Code: * Example generated by -dataex-. To install: s…
Subscribe to:
Post Comments (Atom)
0 Response to Gravity model estimation
Post a Comment