Stepwise Regression specific process

[Giles German]

I have the following data set and want to order the variables by the variables with the least missing. I then want to complete a stepwise regression adding the variables (in that order) with anything with a t stat of <0.2 remaining plus also a mean vif of <1.5.

I can't manage to include the mean vif and I can't get the stepwise to regress in the order of the variable with the least missing first either. The dependent variable is the WeekPriceChangePercent.

Giles

* Example generated by -dataex-. To install: ssc install dataex
clear
input double(WeekPriceChangePercent PricetoBookValueMRQ PricetoEarningsRatio Beta5year BookValueperShareMRFY CashperShareTTM CurrentRatio CapExGrowthRate5Year)
 .1366  2.84  17.31   1.59   4.33  .0088  2.15  .1439
     0  2.36  22.51  .2313   23.1  .0336  2.73   .298
 .4236  11.1 114.09 -.8464   5.56  .0173 21.44 1.1844
-.1139  4.44  22.05 -.0624   1.21  .0032  1.16 -.1491
 -.176  1.34  18.35  .3799  13.11  .0119     . -.0331
 .2263  1.09      .   1.49   8.37  .0031   1.1  .1901
2.9762 12.27      .   3.09  .2683 -.0014   5.7 -.0153
 .0322  1.66  22.55  .4906  14.38  .0216  2.04  .0359
1.0394  9.78      .  -.168   4.14 -.0032  3.49  .3163
 .3832  9.94      .   1.56   1.09  -.008  4.09  -.238
 .2045  1.14   3.94  .2252  .7341   .001  4.56      .
-.1926 .8963  11.31   1.06  .3949  .0005  3.53      .
 -.285  2.06      .   1.61   5.54 -.0523  2.63  .4073
 .3249  1.46   7.41   1.59   8.24      .  4.15 -.1294
 .3778 11.99  27.29   1.06  17.26  .1034  2.07  .0059
 .4126  6.35      .   1.44  .1962  .0005 .3203  .4519
1.2924  5.42  46.34      .  .3966  .0004  1.45  .5458
-.5089  2.97      . -.0126   1.83 -.0017  8.44      .
 .0659 25.85  26.82  .8692   1.04  .0088  1.44  .0269
  .648  1.99  19.95  .4067  .7695 -.0028  2.41      .
 .4341  1.99  19.95  .4067  .7695 -.0028  2.41      .
 .0866 .4649  13.44   1.35  41.43  .0194     .  .0731
-.0049  1.36  15.06      .    1.4  .0045  1.32 -.0011
 .2855 14.98 129.42    1.4  .3324  .0011 .7228  .1564
 -.029  4.61   35.3      .    2.5  .0004  3.26      .
-.0097  4.57      .  .5491    2.3  .0005  3.94  .4425
 .0967  1.09  10.95  .7033   4.69   .004 .5064      .
 .0732  1.09  10.95  .7033   4.69   .004 .5064      .
-.1144  3.96  18.57  .8304   4.68      . .1066      .
 .8008   1.2  17.52  .9103   4.69   .006  1.31  .0935
 .2911  4.21  21.95  .6062   1.56   .004  1.35  .1232
-.0965  5.96      .   1.06  .9236 -.0021  2.06  .0614
2.3113 23.15  62.61  .9738  .1755  .0002  2.52 -.2459
 .1956  1.68  35.59   1.01  .7783  .0012  1.37  .0733
 .1828 .8827  54.96   1.22  .0666      0  5.44 -.2133
 -.422     .   9.69  .7371  -3.14  .0025 .6413   .217
-.3402     .   9.69  .7371  -3.14  .0025 .6413   .217
 .1588  1.31      .      .   5.27  .0034  3.28      .
 .9641     .  29.74  .0775    1.3      .     .  .2633
 .7186     .  29.74  .0775    1.3      .     .  .2633
-.1418 .6537  32.62      .   .163      0  7.21      .
     . .6537  32.62      .   .163      0  7.21      .
   .21  4.23  27.59    1.3   16.1  .0312  1.97  .2279
 .2118  4.23  27.59    1.3   16.1  .0312  1.97  .2279
 .4597  3.53  25.48   .943  11.02  .0241  1.28  .0651
 .5766  3.53  25.48   .943  11.02  .0241  1.28  .0651
     .  3.53  25.48   .943  11.02  .0241  1.28  .0651
 .1269  1.93   16.2   1.16  53.35  .1245 .9645 -.0009
 .3069   1.6  91.29   1.41  19.38  .0198  2.65  -.181
 .2691  1.85  21.72  .2069   15.1  .1501     . -.0598
 .1162  1.85  21.72  .2069   15.1  .1501     . -.0598
  .609  6.07  44.04  .8098   1.48  .0024  3.42 1.0824
-.1408 .7159   7.27  .8573   8.73  .0104  1.58 -.0028
  -.32 17.41      .   1.68 -.1083  -.007  3.32  .0389
-.3438 17.41      .   1.68 -.1083  -.007  3.32  .0389
 .1207  3.27  19.19   1.02   5.17  .0127  2.32  .0276
 .1223  3.27  19.19   1.02   5.17  .0127  2.32  .0276
  .278  1.23   7.54  .6028   1.77   .002  4.12 -.0221
 .0984  5.11  52.17   1.02    8.6  .0133  6.34  .0583
 .2619  4.21   26.8   1.06   4.56  .0098  1.37 -.0403
 .1431  4.17  26.53   1.14   4.56  .0098  1.37 -.0403
 .2128  4.17  26.53   1.14   4.56  .0098  1.37 -.0403
 .1193  1.02   6.02  .3617  12.77  .0213  4.29    .07
  .445   3.2  48.51   1.53  10.16  .0117  2.69 -.0554
 .6374 23.95  24.37   1.52   2.12  .0321  1.45  .0614
-.0238  2.44  15.47  .1461   2.45  .0048  8.23      .
  .063   2.4  19.59  .4256  17.44  .0262  5.57 -.0223
 .3958  7.44  53.91  .7319    1.5  .0024   4.4   .466
 .9073  4.91  24.64      .   8.53  .0093  1.44      .
1.9346  7.15      .   1.28   1.79 -.0036 22.13  .6696
 .6159  1.12      .  .9933  13.37  .0231  2.08 -.1506
 .3949  3.52  20.87  .6387   3.49 -.0021   1.1  .1646
 .1768  3.72  12.36  .4359  .1745  .0003  1.25 -.3412
-.0241  3.72  12.36  .4359  .1745  .0003  1.25 -.3412
     .  8.82      .      .   3.15      .  1.01      .
 .4141   4.4  21.87  .8585   7.92  .0168   3.3 1.1912
     .   1.1   7.52      .  .0467      .  2.73      .
 1.978 27.72      .   1.94    1.5 -.0024  4.09  .0348
-.0226  1.85  27.96  .7637   15.3  .0153  1.63  .1536
-.1401  1.85  27.96  .7637   15.3  .0153  1.63  .1536
 .2851  1.26  11.72      .  17.88  .0079     .      .
 .2728  1.26  11.72      .  17.88  .0079     .      .
-.2609     .      .  .5346      0      .     1      .
-.0238  5.31     20      .   1.71  .0014  2.79  .1994
 .0266     .      .  .7984 -.9886  .0036 .6058   .283
-.5283     .   8.53  .4657   3.31  .0007     . -.0665
-.5378     .   8.53  .4657   3.31  .0007     . -.0665
-.2115  1.36  35.45  .5598  13.84  .0169     .  .4084
 .0946  1.21   14.8  .3462   9.36   .008  2.21  .4445
 .0511   2.7  17.55   .827  .9977  .0016  1.15  .0843
 .3111 .7359  12.07   .365   1.03  .0038  3.24 -.1375
 .2941 .7359  12.07   .365   1.03  .0038  3.24 -.1375
-.4444 .6802      .      . -.3647      . .9193      .
 .0644 11.66      .   2.17    1.1 -.0089 21.23      .
 .1563  1.96  25.66  .8887  10.97  .0157  1.77  .1089
 .1561  1.96  25.66  .8887  10.97  .0157  1.77  .1089
-.3759  1.31  16.44   1.05  .3991  .0005  1.43  .5511
-.0617  1.41  18.44  .6746   4.85 -.0012  3.03      .
 .3829 11.17  28.37   1.03  10.25  .0577  1.25  .0676
 .7031  1.64  73.76   1.38   4.96  .0002 22.69  .2771
end

My effort

preserve
drop WeekPriceChangePercent
misstable tree, asis
return list
restore

display r(vars)
order `r(vars)'

local z WeekPriceChangePercent
stepwise, pe(0.2): regress `z' `r(vars)'

Please can you help me?

[Carlo Lazzaro]

Giles:
obviously with no fault from your side, the subject of your question is not that popular on this forum (see: https://www.stata.com/support/faqs/s...sion-problems/): that's why you've not received any reply, so far.
That said, I would not sponsor your approach of focusing (if I get you right) your analysis on the set of variables with least missing values, unless you have already checked that their missingness in not informative. Skipping this step would add another bias to your research (in addition to the one underlying any -stepwise- procedure).

[William Lisowski]

Carlo accidentally copied-and-pasted an abbreviated version of the link he included in his post. The following should work.

https://www.stata.com/support/faqs/s...sion-problems/

I agree with Carlo, both as to his concerns and his recommendation of this link.

[Giles German]

Thanks Carlo I have a plan consider this closed.

[Carlo Lazzaro]

William is right:
from this side of the pound, copy and paste does not seem to work properly on this fading-away Sunday!
As an aside, I think that Giles made a wise decision in devoting his research energies to other statistical approaches.