hi,
i've run treatreg using stata as well as a 2-step model to determine impact
of the choice variable of technology adoption (phatpot2b or pot2) on the
income per hectare (profitha). With the second step, this is what i got:
Source | SS df MS Number of obs = 587
-------------+------------------------------ F( 21, 565) =
15.54
Model | 367.309377 21 17.4909227 Prob > F =
0.0000
Residual | 635.851023 565 1.12540004 R-squared =
0.3662
-------------+------------------------------ Adj R-squared =
0.3426
Total | 1003.1604 586 1.71187782 Root MSE =
1.0608
----------------------------------------------------------------------------
--
lprofitha | Coef. Std. Err. t P>|t| [95% Conf.
Interval]
-------------+--------------------------------------------------------------
--
larea | -.3422196 .0641317 -5.34 0.000 -.4681853
-.2162539
r9 | .7074985 .2717812 2.60 0.009 .1736737
1.241323
r11 | .449468 .2905934 1.55 0.122 -.1213073
1.020243
lyield | .6401999 .0559619 11.44 0.000 .5302811
.7501187
tenure1 | .3332755 .1328914 2.51 0.012 .0722541
.5942969
tenure2 | .5958305 .1804391 3.30 0.001 .2414171
.9502439
proof | .2982886 .1485377 2.01 0.045 .0065351
.5900421
yield1 | 1.240719 .5314508 2.33 0.020 .1968581
2.284579
arbdum | -.1812507 .107392 -1.69 0.092 -.392187
.0296856
agehdum2 | .2887208 .1028717 2.81 0.005 .0866632
.4907784
phatpot2b | -.2493435 .1585794 -1.57 0.116 -.5608205
.0621336
coopmem | .226677 .1021316 2.22 0.027 .026073
.4272809
labdum | -.228383 .1124028 -2.03 0.043 -.4491614
-.0076046
r0 | .9007858 .280013 3.22 0.001 .3507923
1.450779
r1 | .8929879 .1844589 4.84 0.000 .5306789
1.255297
r2 | .6005806 .1391465 4.32 0.000 .3272731
.8738881
r3 | .6606746 .1849266 3.57 0.000 .2974469
1.023902
r4 | .7185069 .2064196 3.48 0.001 .3130634
1.12395
r5 | .5458338 .1690473 3.23 0.001 .2137958
.8778717
r7 | -.4041432 .226489 -1.78 0.075 -.8490064
.0407201
r8 | .4669252 .1860029 2.51 0.012 .1015836
.8322668
_cons | 7.555088 .2484353 30.41 0.000 7.067118
8.043057
The adjusted R2 is not that high. Does this mean i won't be able to use
this model to interpret the relationship of the choice variable to the
dependent variable, profitha?
Second question is related to this.Since the fit of the profit equation is
not that large judging by the value of the adjusted R2, does this mean I
can't proceed to analyzing the covariance matrix between the error terms in
the first and second steps? here's the treatreg output:
Treatment effects model -- MLE Number of obs =
221
Wald chi2(9) =
57.10
Log likelihood = -2404.7144 Prob > chi2 =
0.0000
----------------------------------------------------------------------------
--
| Coef. Std. Err. z P>|z| [95% Conf.
Interval]
-------------+--------------------------------------------------------------
--
profitha |
area | -979.9762 563.985 -1.74 0.082 -2085.367
125.4141
yield | 1878.773 412.2671 4.56 0.000 1070.744
2686.802
yield1 | -4060.235 4292.129 -0.95 0.344 -12472.65
4352.184
arcdum | 1985.837 1865.546 1.06 0.287 -1670.566
5642.239
agehdum | 3572.093 2728.362 1.31 0.190 -1775.397
8919.584
distdum | 1531.586 1521.883 1.01 0.314 -1451.25
4514.422
extn | 2109.938 2019.068 1.05 0.296 -1847.364
6067.239
labdum2 | 14748.99 4642.952 3.18 0.001 5648.969
23849.01
pot2 | -11077.05 3577.151 -3.10 0.002 -18088.14
-4065.96
_cons | 7995.068 3541.648 2.26 0.024 1053.565
14936.57
-------------+--------------------------------------------------------------
--
pot2 |
area | .1083771 .1352427 0.80 0.423 -.1566937
.3734479
irrig | 2.206678 .5429927 4.06 0.000 1.142432
3.270924
educhdum | -1.380165 .3418587 -4.04 0.000 -2.050196
-.7101345
dist | -.6386896 .2883079 -2.22 0.027 -1.203763
-.0736164
arcdum | .2979616 .566744 0.53 0.599 -.8128362
1.408759
arbdum | .252918 .3448144 0.73 0.463 -.4229058
.9287417
ageh | .0102057 .0140028 0.73 0.466 -.0172393
.0376507
extn | .4007376 .3768061 1.06 0.288 -.3377888
1.139264
credit | .4109917 .367625 1.12 0.264 -.30954
1.131523
tenure2 | -.5366819 .421654 -1.27 0.203 -1.363108
.2897447
_cons | .1249916 .9564531 0.13 0.896 -1.749622
1.999605
-------------+--------------------------------------------------------------
--
/athrho | .56256 .2833076 1.99 0.047 .0072873
1.117833
/lnsigma | 9.315738 .0499618 186.46 0.000 9.217814
9.413661
-------------+--------------------------------------------------------------
--
rho | .5098743 .2096556 .0072872
.8068137
sigma | 11111.52 555.1518 10075.02
12254.65
lambda | 5665.479 2433.118 896.6554
10434.3
----------------------------------------------------------------------------
--
LR test of indep. eqns. (rho = 0): chi2(1) = 4.16 Prob > chi2 =
0.0414
----------------------------------------------------------------------------
--
hope you can enlighten me on this:) thanks a lot,
aileen
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
