Stata Journals 2013 Volume 13 No.1 中提供的命令
具体可以看帮助或者到论坛下载sj13-1

reganat --   Graphical inspection of linear multivariate models

Syntax

        reganat depvar varlist [if] [in] [, options]

    options               Description
    --------------------------------------------------------------------------------------------------------------------------------
    Options
      dis(varlist)        graphs only the variables in varlist and omits the rest
      biline              plots a regression line for the bivariate linear model
      biscat              plots a scatterplot for the bivariate linear model
      reg                 displays the results for the estimation of the multivariate model
      nolegend            prevents the legend to be displayed
      nocovlist           prevents the list of covariates to be displayed
      scheme(scheme)      specifies the graphical scheme
    --------------------------------------------------------------------------------------------------------------------------------
    by is not allowed.


Description

    reganat is a graphical tool for inspecting the effect of a covariate on a dependent variable in the context of multivariate OLS   estimation.  The name is an acronym for the expression regression anatomy, a result in OLS' algebra originally due to Frisch and    Waugh (1933) and recently revived by Angrist and Pischke's Mostly Harmless Econometrics (2009).  
     In a bivariate regression model  Y = bx1 + g, the graphical inspection of the scatterplot provides useful information on the relation between the independent  variable x1 and the dependent variable Y, but can be highly misleading when the underlying real model is multivariate of the type Y = X'B + e where X' includes also x1.  
      In general, the OLS multivariate estimator is not equivalent to an OLS estimator obtained using a separate regression on each independent variable since correlation among independent variables must be accounted for.  
      Angrist and Pischke (2009) show that in a multivariable model, the regression parameter for a given regressor is the bivariate slope coefficient for the corresponding regressor after partialling out all other covariates.  Accordingly, this  command displays a table of scatterplots, with the dependent variable plotted against the independent variable net of any linear   correlation with the other independent variables.  This combined graph can be helpful when inspecting the data for outliers, nonlinearities, and other modelling issues.



********************************************************************
*例子：
*Setup
sysuse auto, clear

*Obtain a combined graph of the effect of several regressors
reganat price length weight headroom mpg,reg

*Obtain a combined graph of the effect of a subset of the regressors,
* along with scatterplots and fitted line for the univariate     models
reganat price length weight headroom mpg, dis(weight length) biline  reg


*******************************************

. *Setup
. sysuse auto, clear
(1978 Automobile Data)

.
. *Obtain a combined graph of the effect of several regressors
. reganat price length weight headroom mpg,reg
Dependent variable: price
Independent variables: length weight headroom mpg
Plotting: length weight headroom mpg

      Source |       SS       df       MS              Number of obs =      74
-------------+------------------------------           F(  4,    69) =   10.21
       Model |   236190226     4  59047556.6           Prob > F      =  0.0000
    Residual |   398875170    69  5780799.56           R-squared     =  0.3719
-------------+------------------------------           Adj R-squared =  0.3355
       Total |   635065396    73  8699525.97           Root MSE      =  2404.3

------------------------------------------------------------------------------
       price |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
      length |  -94.49651   40.39563    -2.34   0.022    -175.0836   -13.90944
      weight |   4.335045   1.162745     3.73   0.000     2.015432    6.654657
    headroom |  -490.9667   388.4892    -1.26   0.211    -1265.981     284.048
         mpg |  -87.95838    83.5927    -1.05   0.296    -254.7213    78.80449
       _cons |   14177.58   5872.766     2.41   0.018     2461.735    25893.43
------------------------------------------------------------------------------

.
. *Obtain a combined graph of the effect of a subset of the regressors,
. * along with scatterplots and fitted line for the univariate     models
. reganat price length weight headroom mpg, reg dis(weight length) biline
Dependent variable: price
Independent variables: length weight headroom mpg
Plotting: weight length

      Source |       SS       df       MS              Number of obs =      74
-------------+------------------------------           F(  4,    69) =   10.21
       Model |   236190226     4  59047556.6           Prob > F      =  0.0000
    Residual |   398875170    69  5780799.56           R-squared     =  0.3719
-------------+------------------------------           Adj R-squared =  0.3355
       Total |   635065396    73  8699525.97           Root MSE      =  2404.3

------------------------------------------------------------------------------
       price |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
      length |  -94.49651   40.39563    -2.34   0.022    -175.0836   -13.90944
      weight |   4.335045   1.162745     3.73   0.000     2.015432    6.654657
    headroom |  -490.9667   388.4892    -1.26   0.211    -1265.981     284.048
         mpg |  -87.95838    83.5927    -1.05   0.296    -254.7213    78.80449
       _cons |   14177.58   5872.766     2.41   0.018     2461.735    25893.43
------------------------------------------------------------------------------