I am using a bilateral exchange rate variable in the gravity model. However, when I use the exchange rate variables in the log form, it changes the overall results mainly with opposite sign. However, using without log make the thing in the right way. Please let me know what are reasons and how to deal with it. The descriptive statistics and regression are appended below.
Thanks.
sum exc lexc
PHP Code:
Variable | Obs Mean Std. Dev. Min Max
-------------+--------------------------------------------------------
exc | 78176 591.1251 3470.122 .0004463 115588.5
lexc | 78176 1.843932 3.34576 -7.714482 11.65779
xtreg lremit $xlist1 i.year, vce(cluster pairid) fe
PHP Code:
Fixed-effects (within) regression Number of obs = 1089
Group variable: pairid Number of groups = 267
R-sq: within = 0.1237 Obs per group: min = 1
between = 0.4856 avg = 4.1
overall = 0.5531 max = 7
F(11,266) = 5.67
corr(u_i, Xb) = 0.0832 Prob > F = 0.0000
(Std. Err. adjusted for 267 clusters in pairid)
------------------------------------------------------------------------------
| Robust
lremit | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
lgdpc | .3832628 .1523686 2.52 0.012 .0832609 .6832647
lgdpc_hos | .2996346 .3462619 0.87 0.388 -.3821282 .9813974
lcost2 | -.1845849 .0986203 -1.87 0.062 -.3787607 .0095908
lmig_st | .3717788 .1603001 2.32 0.021 .0561603 .6873973
exc | -.0001012 .0000531 -1.91 0.058 -.0002058 3.40e-06
|
year |
2012 | .0051369 .0194579 0.26 0.792 -.0331742 .043448
2013 | -.02299 .0473106 -0.49 0.627 -.1161409 .0701608
2014 | -.0251942 .0545859 -0.46 0.645 -.1326696 .0822811
2015 | .1554102 .0574091 2.71 0.007 .0423762 .2684442
2016 | .0830776 .0632646 1.31 0.190 -.0414855 .2076407
2017 | .1111879 .0632979 1.76 0.080 -.0134407 .2358166
|
_cons | -7.122939 5.249512 -1.36 0.176 -17.45882 3.212942
-------------+----------------------------------------------------------------
sigma_u | 1.2772635
sigma_e | .34175982
rho | .93318862 (fraction of variance due to u_i)
-----------------------------------------------------------------------------
xtreg lremit $xlist1 i.year, vce(cluster pairid) fe
PHP Code:
Fixed-effects (within) regression Number of obs = 1089
Group variable: pairid Number of groups = 267
R-sq: within = 0.2208 Obs per group: min = 1
between = 0.1570 avg = 4.1
overall = 0.1428 max = 7
F(11,266) = 6.06
corr(u_i, Xb) = -0.9014 Prob > F = 0.0000
(Std. Err. adjusted for 267 clusters in pairid)
------------------------------------------------------------------------------
| Robust
lremit | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
lgdpc | 1.104068 .2582936 4.27 0.000 .5955077 1.612628
lgdpc_hos | -.7614625 .3678625 -2.07 0.039 -1.485755 -.0371698
lcost2 | -.1180805 .0793946 -1.49 0.138 -.2744023 .0382413
lmig_st | .5051947 .1796623 2.81 0.005 .1514536 .8589359
lexc | 1.283118 .3861076 3.32 0.001 .5229025 2.043334
|
year |
2012 | -.077954 .0267472 -2.91 0.004 -.1306171 -.0252908
2013 | -.1662693 .0760294 -2.19 0.030 -.3159652 -.0165734
2014 | -.2275463 .0978052 -2.33 0.021 -.4201172 -.0349753
2015 | -.0994765 .0704605 -1.41 0.159 -.2382079 .0392548
2016 | -.2454423 .0879119 -2.79 0.006 -.4185339 -.0723506
2017 | -.2453342 .1058131 -2.32 0.021 -.453672 -.0369964
|
_cons | -5.906389 4.652194 -1.27 0.205 -15.0662 3.25342
-------------+----------------------------------------------------------------
sigma_u | 3.962099
sigma_e | .32225379
rho | .99342824 (fraction of variance due to u_i)
------------------------------------------------------------------------------
0 Response to Bilateral exchange rate data with and without make changes in overall results.
Post a Comment