-------------------------------------------------------------------------------
       log:  A:\class18.log
  log type:  text
 opened on:  12 Nov 2002, 12:55:20

. use a:\class18.dta

. summarize

    Variable |     Obs        Mean   Std. Dev.       Min        Max
-------------+-----------------------------------------------------
        size |     100        4.35   1.493251          2         10
       earns |     100       7.911   5.129704        .75         30
      income |     100     9.94124   5.583998        .75      32.08
      wealth |     100    12.63536   16.81237        .05     98.033
      saving |     100     1.86691   4.447998     -5.577     31.599
          id |     100        50.5   29.01149          1        100

. regress saving income wealth

      Source |       SS       df       MS              Number of obs =     100
-------------+------------------------------           F(  2,    97) =    8.07
       Model |  279.445984     2  139.722992           Prob > F      =  0.0006
    Residual |  1679.23772    97   17.311729           R-squared     =  0.1427
-------------+------------------------------           Adj R-squared =  0.1250
       Total |   1958.6837    99  19.7846839           Root MSE      =  4.1607

------------------------------------------------------------------------------
      saving |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
      income |   .3500154   .1011997     3.46   0.001     .1491621    .5508687
      wealth |  -.0273428   .0336121    -0.81   0.418    -.0940535    .0393678
       _cons |  -1.267191   .8892972    -1.42   0.157      -3.0322    .4978178
------------------------------------------------------------------------------

. generate x1s = income^2

. generate x2s = wealth^2

. generate x1x2 = income*wealth

. hettest income wealth x1s x2s x1x2

Cook-Weisberg test for heteroskedasticity using variables specified
     Ho: Constant variance
         chi2(5)      =    301.08
         Prob > chi2  =      0.0000

. hettest

Cook-Weisberg test for heteroskedasticity using fitted values of saving
     Ho: Constant variance
         chi2(1)      =    109.83
         Prob > chi2  =      0.0000

. regress saving income wealth, robust

Regression with robust standard errors                 Number of obs =     100
                                                       F(  2,    97) =    2.47
                                                       Prob > F      =  0.0897
                                                       R-squared     =  0.1427
                                                       Root MSE      =  4.1607

------------------------------------------------------------------------------
             |               Robust
      saving |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
      income |   .3500154   .2432482     1.44   0.153    -.1327649    .8327958
      wealth |  -.0273428    .061659    -0.44   0.658    -.1497189    .0950333
       _cons |  -1.267191    1.53312    -0.83   0.411     -4.31001    1.775628
------------------------------------------------------------------------------

. log close
       log:  A:\class18.log
  log type:  text
 closed on:  12 Nov 2002, 12:58:03
-------------------------------------------------------------------------------