关于boxcox检验的问题

hamlet1983118

3132

收藏 2009-05-12

. boxcox rmedicalexpenditureper rgorvenmentrevenueper rtransferper, notrans( above
> 65proportion below15proportion morbilityper1000 mortalityper1000 nonprofitbedsp
> er10000 urbanproportion femaleproportion juniorcollegeedupropotion unmarried ins
> urancedcoverage year2003 year2004 year2005 year2006) model(lambda)
Fitting comparison model

Iteration 0:   log likelihood = -899.82201
Iteration 1:   log likelihood = -808.6929
Iteration 2:   log likelihood = -808.2444
Iteration 3:   log likelihood = -808.24417
Iteration 4:   log likelihood = -808.24417

Fitting full model

Iteration 0:   log likelihood = -712.2262
Iteration 1:   log likelihood = -644.48226
Iteration 2:   log likelihood = -640.36162
Iteration 3:   log likelihood = -623.09097
Iteration 4:   log likelihood = -622.92809
Iteration 5:   log likelihood = -622.92764
Iteration 6:   log likelihood = -622.92764

                                                  Number of obs   =        155
                                                  LR chi2(16)     =     370.63
Log likelihood = -622.92764                       Prob > chi2     =      0.000

------------------------------------------------------------------------------
rmedicalex~r |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     /lambda |   .0405063    .073433     0.55   0.581    -.1034199    .1844324
------------------------------------------------------------------------------

Estimates of scale-variant parameters
----------------------------
             |      Coef.
-------------+--------------
Notrans      |
above65pro~n | -5.976282
below15pro~n |    .485161
morbili~1000 |   .0273048
mortali~1000 |   10.51786
nonpro~10000 | -.0041632
urbanpropo~n | -1.656319
femaleprop~n |   10.72626
juniorcoll~n |   3.717347
   unmarried |   1.858071
insuranced~e | -.8560323
    year2003 |    .053958
    year2004 | -.0405212
    year2005 |   .0450074
    year2006 |   .0511788
       _cons | -9.382694
-------------+--------------
Trans        |
rgorvenmen~r |   .7286858
rtransferper |   .4800122
-------------+--------------
      /sigma |   .2112387
----------------------------

---------------------------------------------------------
   Test         Restricted     LR statistic      P-value
    H0:       log likelihood       chi2       Prob > chi2
---------------------------------------------------------
lambda = -1     -677.88107       109.91           0.000
lambda = 0     -623.07519         0.30           0.587
lambda = 1      -712.2262       178.60           0.000
---------------------------------------------------------
上面是我做的boxcox检验。