Stata 计算平均数差

变量

bwnt

给出婴儿的出生体重（以盎司为单位）

吸烟母亲和不吸烟母亲的平均出生体重有什么区别

---

如何在Stata中进行均值减法？

以下是使用

auto

toy数据集进行减法的一种方法：

sysuse auto, clear

summarize price if foreign
local mean1 = r(mean)

summarize price if !foreign
local mean2 = r(mean)

如果您只想查看结果，可以使用

display

命令：

display `mean1' - `mean2'
312.25874

generate mean_price = `mean1' - `mean2'

list mean_price in 1, abbreviate(10)

     +------------+
     | mean_price |
     |------------|
  1. |   312.2587 |
     +------------+

regress price foreign 


      Source |       SS           df       MS      Number of obs   =        74
-------------+----------------------------------   F(1, 72)        =      0.17
       Model |  1507382.66         1  1507382.66   Prob > F        =    0.6802
    Residual |   633558013        72  8799416.85   R-squared       =    0.0024
-------------+----------------------------------   Adj R-squared   =   -0.0115
       Total |   635065396        73  8699525.97   Root MSE        =    2966.4

------------------------------------------------------------------------------
       price |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     foreign |   312.2587   754.4488     0.41   0.680    -1191.708    1816.225
       _cons |   6072.423    411.363    14.76   0.000     5252.386     6892.46
------------------------------------------------------------------------------

webuse fuel3            (setup)
ttest mpg, by(treated)  (two-sample t test using groups)

如果要将结果保存在变量中，可以使用

generate

命令：

display `mean1' - `mean2'
312.25874

generate mean_price = `mean1' - `mean2'

list mean_price in 1, abbreviate(10)

     +------------+
     | mean_price |
     |------------|
  1. |   312.2587 |
     +------------+

regress price foreign 


      Source |       SS           df       MS      Number of obs   =        74
-------------+----------------------------------   F(1, 72)        =      0.17
       Model |  1507382.66         1  1507382.66   Prob > F        =    0.6802
    Residual |   633558013        72  8799416.85   R-squared       =    0.0024
-------------+----------------------------------   Adj R-squared   =   -0.0115
       Total |   635065396        73  8699525.97   Root MSE        =    2966.4

------------------------------------------------------------------------------
       price |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     foreign |   312.2587   754.4488     0.41   0.680    -1191.708    1816.225
       _cons |   6072.423    411.363    14.76   0.000     5252.386     6892.46
------------------------------------------------------------------------------

webuse fuel3            (setup)
ttest mpg, by(treated)  (two-sample t test using groups)

---

您也可以使用

mean

命令，而不是

summary

：

mean price, over(foreign)
matrix A = r(table)

display A[1,2] - A[1,1]
312.25874

generate mean_price = A[1,2] - A[1,1]

list mean_price in 1, abbreviate(10)

     +------------+
     | mean_price |
     |------------|
  1. |   312.2587 |
     +------------+

---

编辑：

尼克·考克斯的评论

---

您还可以使用

回归

命令：

display `mean1' - `mean2'
312.25874

generate mean_price = `mean1' - `mean2'

list mean_price in 1, abbreviate(10)

     +------------+
     | mean_price |
     |------------|
  1. |   312.2587 |
     +------------+

regress price foreign 


      Source |       SS           df       MS      Number of obs   =        74
-------------+----------------------------------   F(1, 72)        =      0.17
       Model |  1507382.66         1  1507382.66   Prob > F        =    0.6802
    Residual |   633558013        72  8799416.85   R-squared       =    0.0024
-------------+----------------------------------   Adj R-squared   =   -0.0115
       Total |   635065396        73  8699525.97   Root MSE        =    2966.4

------------------------------------------------------------------------------
       price |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     foreign |   312.2587   754.4488     0.41   0.680    -1191.708    1816.225
       _cons |   6072.423    411.363    14.76   0.000     5252.386     6892.46
------------------------------------------------------------------------------

webuse fuel3            (setup)
ttest mpg, by(treated)  (two-sample t test using groups)

---

均值的差异是变量

foreign

的系数，之后可以使用

\u b[foreign]

访问该系数。对于代码相差1的任何二进制预测器（例如，

0

和

1

，

1

和

2

）也是如此。

您也可以使用

ttest

命令：

display `mean1' - `mean2'
312.25874

generate mean_price = `mean1' - `mean2'

list mean_price in 1, abbreviate(10)

     +------------+
     | mean_price |
     |------------|
  1. |   312.2587 |
     +------------+

regress price foreign 


      Source |       SS           df       MS      Number of obs   =        74
-------------+----------------------------------   F(1, 72)        =      0.17
       Model |  1507382.66         1  1507382.66   Prob > F        =    0.6802
    Residual |   633558013        72  8799416.85   R-squared       =    0.0024
-------------+----------------------------------   Adj R-squared   =   -0.0115
       Total |   635065396        73  8699525.97   Root MSE        =    2966.4

------------------------------------------------------------------------------
       price |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     foreign |   312.2587   754.4488     0.41   0.680    -1191.708    1816.225
       _cons |   6072.423    411.363    14.76   0.000     5252.386     6892.46
------------------------------------------------------------------------------

webuse fuel3            (setup)
ttest mpg, by(treated)  (two-sample t test using groups)

对你来说，那就是：

ttest bwght, by(smoking)

---

如果你发现我的答案有用，请考虑使用上箭头投票。