Question

我有以下模型：

ivreg ldemand social_housing transport year (lprice = utilities)

但是，我想将year作为虚拟变量。

我如何在Stata中做到这一点？

使用i.year不能用于ivreg命令。

交叉发布于Statalist。

Answer 1

命令ivreg不允许使用因子变量：

. webuse hsng2, clear

. ivreg rent pcturban i.region (hsngval = faminc)
factor variables not allowed
r(101);

但是，您仍然可以使用xi前缀动态创建虚拟对象：

.  xi: ivreg rent pcturban  i.region (hsngval = faminc)
i.region          _Iregion_1-4        (naturally coded; _Iregion_1 omitted)

Instrumental variables (2SLS) regression

      Source |       SS           df       MS      Number of obs   =        50
-------------+----------------------------------   F(5, 44)        =      9.10
       Model |  12735.4667         5  2547.09334   Prob > F        =    0.0000
    Residual |  48507.6533        44  1102.44667   R-squared       =    0.2079
-------------+----------------------------------   Adj R-squared   =    0.1179
       Total |    61243.12        49  1249.85959   Root MSE        =    33.203

------------------------------------------------------------------------------
        rent |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     hsngval |   .0038683   .0008958     4.32   0.000     .0020629    .0056737
    pcturban |  -.4980121   .5179779    -0.96   0.342    -1.541928    .5459039
  _Iregion_2 |   1.528672   15.14086     0.10   0.920    -28.98572    32.04306
  _Iregion_3 |    7.74279   15.10906     0.51   0.611    -22.70752     38.1931
  _Iregion_4 |  -40.61235   19.60999    -2.07   0.044    -80.13369   -1.091002
       _cons |   88.26681   31.69154     2.79   0.008     24.39671    152.1369
------------------------------------------------------------------------------
Instrumented:  hsngval
Instruments:   pcturban _Iregion_2 _Iregion_3 _Iregion_4 faminc
------------------------------------------------------------------------------

重要的是要注意，根据命令的帮助文件：

过期命令

ivreg 是Stata 10以来的过期命令。 ivreg 已由 ivregress 命令替换。

因此，最好改用ivregress：

. ivregress 2sls rent pcturban i.region (hsngval = faminc), small

Instrumental variables (2SLS) regression

      Source |       SS       df       MS         Number of obs   =         50
-------------+------------------------------      F(  5,    44)   =       9.10
       Model |  12735.4667     5  2547.09334      Prob > F        =     0.0000
    Residual |  48507.6533    44  1102.44667      R-squared       =     0.2079
-------------+------------------------------      Adj R-squared   =     0.1179
       Total |    61243.12    49  1249.85959      Root MSE        =     33.203

------------------------------------------------------------------------------
        rent |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     hsngval |   .0038683   .0008958     4.32   0.000     .0020629    .0056737
    pcturban |  -.4980121   .5179779    -0.96   0.342    -1.541928    .5459039
             |
      region |
    N Cntrl  |   1.528672   15.14086     0.10   0.920    -28.98572    32.04306
      South  |    7.74279   15.10906     0.51   0.611    -22.70752     38.1931
       West  |  -40.61235   19.60999    -2.07   0.044    -80.13369   -1.091002
             |
       _cons |   88.26681   31.69154     2.79   0.008     24.39671    152.1369
------------------------------------------------------------------------------
Instrumented:  hsngval
Instruments:   pcturban 2.region 3.region 4.region faminc

在Stata的命令提示符下键入help ivregress，以获取更多详细信息。

如何在ivreg模型中包含虚拟变量？

1 个答案: