Linking STATA and GAMS

Uwe A. Schneider

last updated January 2010

Other GAMS interfaces  from/to:  ShadeMap   Gnuplot   Stata    Powerpoint    Html    Awk

 

 

1) Generate a .do file for use in STATA.

Use the option outreg in STATA to create a text file which contains the regression coefficients. For example:

xtgls  PAha  ltemp_season lprecipitation_annual lmrevenue if  pestF==3 & cropgr==1
mfx
outreg using "stata_CrT1_loglog.txt", se bracket nolabel ctitle(pestF3) append

2) Execute STATA

With the option outreg, the STATA output file will look like this:

        (1)     (2)     (3)     (4)     (5)     (6)     (7)     (8)     (9)     (10)    (11)    (12)
        pestF2  pestF3  pestF5  pestF6  pestF7  pestF8  pestF9  pestF10 pestF16 pestF17 pestF18 pestF20
ltemp_season_C  -0.000  0.226   -0.001  0.025   -10.117 0.051   -0.091  5.582   0.005   -0.068  -0.088
        [0.002] [0.147] [0.001] [0.037] [11.823]        [0.203] [0.017]**       [3.240] [0.011] [0.029]*
lprecipitation_annual_mm        -0.001  -0.007  -0.000  0.003   4.250   -0.022  0.062   0.290   -0.002
        [0.000] [0.019] [0.000] [0.004] [4.358] [0.022] [0.008]**       [0.311] [0.004] [0.003] [0.009]
lmrevenue       0.000   0.018   -0.000  0.002   5.724   -0.084  0.014   1.025   -0.001  0.005   -0.004
        [0.001] [0.018] [0.000] [0.003] [3.913] [0.018]**       [0.002]**       [0.338]**       [0.001]
ltotal_area     0.003   0.016   -0.000  -0.003  3.192   -0.051  -0.104  0.402   -0.000  -0.003  -0.007
        [0.001]**       [0.013] [0.000] [0.002] [2.639] [0.016]**       [0.014]**       [0.217] [0.002]
lprecipM        -0.002  0.047   -0.000  -0.002  9.290   -0.072  0.417   0.407   0.004   -0.001  0.036
        [0.001]*        [0.029] [0.000] [0.007] [13.858]        [0.038] [0.054]**       [0.568] [0.006]
lTempavr                -0.610  0.005   -0.069  -4.132  0.513   -1.672  -19.465 -0.029  0.103   0.573
                [0.252]*        [0.003] [0.065] [12.819]        [0.407] [0.224]**       [5.936]**
Constant        -0.012  1.572   -0.014  0.234   -58.319 -1.065  6.117   53.775  0.106   -0.193  -2.040
        [0.009] [0.678]*        [0.007]*        [0.174] [71.403]        [1.172] [0.818]**       [15.926]
Observations    9       52      14      102     10      56      8       67      32      64      24
Number of cropyear      5       13      8       15      5       15      8       15      11      14
Standard errors in brackets
* significant at 5%; ** significant at 1%

3) Translate STATA output file in GAMS format using an awk script.

An example is here. The translated regression data in GAMS format look as below:

Table CrT1_loglog_RegrData(RegressionCoefficients,RegressionItem,Chemical_FamilyCode)
                                             pestF2      pestF3      pestF5      pestF6
          ltemp_season_C   .Coefficient           0       0.226      -0.001       0.025
          ltemp_season_C   .StandardDev       0.002       0.147       0.001       0.037
          ltemp_season_C   .SignifLevel           0           0           0           0
lprecipitation_annual_mm   .Coefficient      -0.001      -0.007           0       0.003
lprecipitation_annual_mm   .StandardDev           0       0.019           0       0.004
lprecipitation_annual_mm   .SignifLevel           0           0           0           0
               lmrevenue   .Coefficient           0       0.018           0       0.002
               lmrevenue   .StandardDev       0.001       0.018           0       0.003
               lmrevenue   .SignifLevel           0           0           0           0
                Constant   .Coefficient      -0.012       1.572      -0.014       0.234
                Constant   .StandardDev       0.009       0.678       0.007       0.174
                Constant   .SignifLevel           0           5           5           0
;

 

Download complete example