stepwise like SPSS? [was: RE: st: Means test]

[Richard Williams <Richard.A.Williams.5@ND.edu>]

Here is a not too painful way to get more stepwise information from 
Stata.  Basically, just run the regressions twice: Once with 
stepwise, and then again with nestreg.  With nestreg, specify the 
vars in the same order that stepwise added them.  Example:

. use "http://www.indiana.edu/~jslsoc/stata/spex_data/ordwarm2.dta";
(77 & 89 General Social Survey)

. sw, pe(.05): reg  warm yr89 male white age ed prst
                      begin with empty model
p = 0.0000 <  0.0500  adding  age
p = 0.0000 <  0.0500  adding  male
p = 0.0000 <  0.0500  adding  yr89
p = 0.0000 <  0.0500  adding  ed
p = 0.0022 <  0.0500  adding  white

[Output deleted]

. nestreg, quietly: reg warm age male yr89 ed white

Block  1: age
Block  2: male
Block  3: yr89
Block  4: ed
Block  5: white

  +-------------------------------------------------------------+
  |       |          Block  Residual                     Change |
  | Block |       F     df        df   Pr > F       R2    in R2 |
  |-------+-----------------------------------------------------|
  |     1 |  103.49      1      2291   0.0000   0.0432          |
  |     2 |   83.64      1      2290   0.0000   0.0769   0.0337 |
  |     3 |   67.28      1      2289   0.0000   0.1033   0.0264 |
  |     4 |   35.46      1      2288   0.0000   0.1170   0.0137 |
  |     5 |    9.44      1      2287   0.0022   0.1206   0.0036 |
  +-------------------------------------------------------------+

You may want to use some of the other nestreg options too, e.g. you 
can ask for likelihood ratio tests rather than Wald and you can have 
it store the results of each model.  And if you want more of the blow 
by blow details, you can leave off the quietly option.  A nice thing 
about the -lr- option is that it also gives you AIC and BIC 
statistics for each model:

. nestreg, quietly lr: reg warm age male yr89 ed white

Block  1: age
Block  2: male
Block  3: yr89
Block  4: ed
Block  5: white

  +----------------------------------------------------------------+
  | Block |        LL       LR     df  Pr > LR       AIC       BIC |
  |-------+--------------------------------------------------------|
  |     1 | -3031.664   101.31      1   0.0000  6067.329  6078.804 |
  |     2 | -2990.536    82.26      1   0.0000  5987.073  6004.286 |
  |     3 | -2957.325    66.42      1   0.0000   5922.65    5945.6 |
  |     4 | -2939.694    35.26      1   0.0000  5889.388  5918.076 |
  |     5 | -2934.973     9.44      1   0.0021  5881.945  5916.371 |
  +----------------------------------------------------------------+


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW (personal):    http://www.nd.edu/~rwilliam
WWW (department):    http://www.nd.edu/~soc 
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/