Hello.
I want to do spatial panel econometric analysis. I'm using gravity model. In the gravitational model as the dependent variable, I used the population, GDP and distance as export explanatory variable. Other than that, I used 3 dummy variables. common border, common language and common belief.
https://www.youtube.com/watch?v=54HoO0KgWLQ I got help from here for codes.
But i face omitted variable. what am i do for this problem?

SDM with spatial fixed-effects Number of obs = 180

Group variable: country Number of groups = 10
Time variable: year Panel length = 18

R-sq: within = 0.9312
between = 0.0494
overall = 0.0039

Mean of fixed-effects = 42.3142

Log-pseudolikelihood = 191.9219
(Std. Err. adjusted for 10 clusters in country)
------------------------------------------------------------------------------
| Robust
v3 | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Main |
v4 | .346386 .1667466 2.08 0.038 .0195686 .6732034
v5 | .080305 .0925599 0.87 0.386 -.101109 .261719
v6 | -2.333392 1.205793 -1.94 0.053 -4.696702 .0299184
v7 | -.1181973 1.267534 -0.09 0.926 -2.602519 2.366124
v8 | 0 (omitted)
v9 | 0 (omitted)
v1 | 0 (omitted)
v2 | 0 (omitted)
-------------+----------------------------------------------------------------
Wx |
v4 | .6347884 .1356332 4.68 0.000 .3689522 .9006246
v5 | .0381044 . . . . .
v6 | -3.933378 2.745495 -1.43 0.152 -9.314448 1.447693
v7 | -.416318 . . . . .
v8 | 0 (omitted)
v9 | 0 (omitted)
v1 | 0 (omitted)
v2 | 0 (omitted)
-------------+----------------------------------------------------------------
Spatial |
rho | .0984262 .1106344 0.89 0.374 -.1184133 .3152656
-------------+----------------------------------------------------------------
Variance |
sigma2_e | .0069144 .0014999 4.61 0.000 .0039746 .0098541
------------------------------------------------------------------------------

. //SDM with spatial fixed-effects [data transformed according to Lee and Yu (20
> 10)]
. xsmle v3 v4 v5 v6 v7 v8 v9 v1 v2, wmat(W) model(sdm) fe type(ind, leeyu) vce(c
> luster country) nolog
Warning: All regressors will be spatially lagged

convergence not achieved

SDM with spatial fixed-effects Number of obs = 170

Group variable: country Number of groups = 10
Time variable: year Panel length = 17

R-sq: within = 0.9312
between = 0.0494
overall = 0.0039

Mean of fixed-effects = 42.3141

Log-pseudolikelihood = 176.4011
(Std. Err. adjusted for 10 clusters in country)
------------------------------------------------------------------------------
| Robust
v3 | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Main |
v4 | .3463859 .1667467 2.08 0.038 .0195683 .6732034
v5 | .0803051 .0925598 0.87 0.386 -.1011088 .2617189
v6 | -2.333389 1.205793 -1.94 0.053 -4.6967 .0299216
v7 | -.4163177 1.267534 -0.33 0.743 -2.90064 2.068004
v8 | 0 (omitted)
v9 | 0 (omitted)
v1 | 0 (omitted)
v2 | 0 (omitted)
-------------+----------------------------------------------------------------
Wx |
v4 | .6347874 .1356332 4.68 0.000 .3689513 .9006236
v5 | .0381044 . . . . .
v6 | -3.933371 2.745496 -1.43 0.152 -9.314444 1.447701
v7 | -.118197 . . . . .
v8 | 0 (omitted)
v9 | 0 (omitted)
v1 | 0 (omitted)
v2 | 0 (omitted)
-------------+----------------------------------------------------------------
Spatial |
rho | .0984269 .1106344 0.89 0.374 -.1184125 .3152662
-------------+----------------------------------------------------------------
Variance |
sigma2_e | .0073211 .0015881 4.61 0.000 .0042084 .0104338
------------------------------------------------------------------------------

. estat ic

Akaike's information criterion and Bayesian information criterion

-----------------------------------------------------------------------------
Model | Obs ll(null) ll(model) df AIC BIC
-------------+---------------------------------------------------------------
. | 170 . 176.4011 8 -336.8022 -311.7159
-----------------------------------------------------------------------------
Note: N=Obs used in calculating BIC; see [R] BIC note.

. //SDM with time fixed-effects
. xsmle v3 v4 v5 v6 v7 v8 v9 v1 v2, wmat(W) model(sdm) fe type(time) vce(cluster
> country) nolog
Warning: All regressors will be spatially lagged