xianfang 发表于 2013-2-19 15:06 
不是的, these two varlists are dependent vars and endogenous var separately.
. webuse hsng2
(1980 Census housing data)
. ivregress 2sls rent pcturban (hsngval popgrow = faminc i.region), small
Instrumental variables (2SLS) regression
Source | SS df MS Number of obs = 50
-------------+------------------------------ F( 3, 46) = 18.47
Model | 21982.9593 3 7327.65311 Prob > F = 0.0000
Residual | 39260.1607 46 853.481754 R-squared = 0.3589
-------------+------------------------------ Adj R-squared = 0.3171
Total | 61243.12 49 1249.85959 Root MSE = 29.214
------------------------------------------------------------------------------
rent | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
hsngval | .0027672 .0005244 5.28 0.000 .0017117 .0038227
popgrow | -.8214817 .4609762 -1.78 0.081 -1.749379 .1064151
pcturban | -.1205201 .4097702 -0.29 0.770 -.9453449 .7043046
_cons | 122.0447 20.08515 6.08 0.000 81.61539 162.474
------------------------------------------------------------------------------
Instrumented: hsngval popgrow
Instruments: pcturban faminc 2.region 3.region 4.region
没有问题的啊