I have run the following regression:
Code:
. logit change firmsize profitability leverage age capitalintensity CAPEX KZindex elektricityge
> nerator Carbonleakage i.twodigitsNACE WestFlanders Hainaut Antwerp Brussels FlemishBrabant Li
> mbourg Liege Namur WalloonBrabant Luxembourg SME publicfirm
note: 16.twodigitsNACE != 0 predicts success perfectly
16.twodigitsNACE dropped and 2 obs not used
note: 21.twodigitsNACE != 0 predicts success perfectly
21.twodigitsNACE dropped and 3 obs not used
note: 25.twodigitsNACE != 0 predicts success perfectly
25.twodigitsNACE dropped and 2 obs not used
note: 28.twodigitsNACE != 0 predicts success perfectly
28.twodigitsNACE dropped and 1 obs not used
note: 30.twodigitsNACE != 0 predicts success perfectly
30.twodigitsNACE dropped and 1 obs not used
note: 42.twodigitsNACE != 0 predicts success perfectly
42.twodigitsNACE dropped and 5 obs not used
note: 47.twodigitsNACE != 0 predicts failure perfectly
47.twodigitsNACE dropped and 1 obs not used
note: 49.twodigitsNACE != 0 predicts success perfectly
49.twodigitsNACE dropped and 2 obs not used
note: 61.twodigitsNACE != 0 predicts failure perfectly
61.twodigitsNACE dropped and 1 obs not used
note: 63.twodigitsNACE != 0 predicts failure perfectly
63.twodigitsNACE dropped and 2 obs not used
note: 70.twodigitsNACE != 0 predicts failure perfectly
70.twodigitsNACE dropped and 1 obs not used
note: 72.twodigitsNACE != 0 predicts success perfectly
72.twodigitsNACE dropped and 1 obs not used
note: 81.twodigitsNACE != 0 predicts success perfectly
81.twodigitsNACE dropped and 1 obs not used
note: Namur != 0 predicts failure perfectly
Namur dropped and 1 obs not used
Iteration 0: log likelihood = -98.836643
Iteration 1: log likelihood = -79.803416
Iteration 2: log likelihood = -79.587861
Iteration 3: log likelihood = -79.566988
Iteration 4: log likelihood = -79.565487
Iteration 5: log likelihood = -79.565147
Iteration 6: log likelihood = -79.565092
Iteration 7: log likelihood = -79.565086
Logistic regression Number of obs = 143
LR chi2(33) = 38.54
Prob > chi2 = 0.2331
Log likelihood = -79.565086 Pseudo R2 = 0.1950
--------------------------------------------------------------------------------------
change | Coef. Std. Err. z P>|z| [95% Conf. Interval]
---------------------+----------------------------------------------------------------
firmsize | .5369456 .2847299 1.89 0.059 -.0211148 1.095006
profitability | -4.871327 6.538349 -0.75 0.456 -17.68626 7.943602
leverage | -2.234129 1.071087 -2.09 0.037 -4.333422 -.1348367
age | .0072455 .0098076 0.74 0.460 -.011977 .026468
capitalintensity | -1.686705 1.344654 -1.25 0.210 -4.322178 .9487684
CAPEX | 1.105687 3.453339 0.32 0.749 -5.662732 7.874106
KZindex | .2451881 .191563 1.28 0.201 -.1302685 .6206447
elektricitygenerator | 12.2858 954.6576 0.01 0.990 -1858.809 1883.38
Carbonleakage | -.625777 .5337667 -1.17 0.241 -1.67194 .4203865
|
twodigitsNACE |
10 | -.2720088 1.277316 -0.21 0.831 -2.775502 2.231484
11 | 2.094524 1.844136 1.14 0.256 -1.519916 5.708964
13 | .2188543 1.574521 0.14 0.889 -2.86715 3.304859
16 | 0 (empty)
17 | -.4607484 1.454024 -0.32 0.751 -3.310583 2.389086
19 | .2352189 1.575869 0.15 0.881 -2.853428 3.323866
20 | .2973133 1.232916 0.24 0.809 -2.119157 2.713784
21 | 0 (empty)
22 | -.5714206 1.640488 -0.35 0.728 -3.786718 2.643877
23 | 1.222801 1.213726 1.01 0.314 -1.156058 3.60166
24 | .8195506 1.376575 0.60 0.552 -1.878486 3.517587
25 | 0 (empty)
28 | 0 (empty)
29 | .0334485 2.098225 0.02 0.987 -4.078997 4.145894
30 | 0 (empty)
35 | -12.92753 954.6579 -0.01 0.989 -1884.023 1858.168
42 | 0 (empty)
46 | .638123 1.392489 0.46 0.647 -2.091105 3.367351
47 | 0 (empty)
49 | 0 (empty)
52 | 1.324916 1.898157 0.70 0.485 -2.395404 5.045236
61 | 0 (empty)
63 | 0 (empty)
70 | 0 (empty)
72 | 0 (empty)
81 | 0 (empty)
|
WestFlanders | -.0334296 .8350943 -0.04 0.968 -1.670184 1.603325
Hainaut | -1.990829 .889228 -2.24 0.025 -3.733684 -.2479741
Antwerp | -1.622771 .7301036 -2.22 0.026 -3.053747 -.1917937
Brussels | -2.43887 1.00253 -2.43 0.015 -4.403793 -.4739474
FlemishBrabant | .5337495 1.402777 0.38 0.704 -2.215643 3.283142
Limbourg | -.6921454 .814151 -0.85 0.395 -2.287852 .9035613
Liege | -1.329429 .944401 -1.41 0.159 -3.180421 .5215631
Namur | 0 (omitted)
WalloonBrabant | -2.447982 1.389671 -1.76 0.078 -5.171687 .2757223
Luxembourg | -1.161801 1.423928 -0.82 0.415 -3.952649 1.629048
SME | 1.970419 .8058729 2.45 0.014 .3909376 3.549901
publicfirm | .1173165 1.683249 0.07 0.944 -3.18179 3.416423
_cons | -8.413368 5.586249 -1.51 0.132 -19.36221 2.535478
--------------------------------------------------------------------------------------Code:
. linktest
Iteration 0: log likelihood = -98.836643
Iteration 1: log likelihood = -80.115266
Iteration 2: log likelihood = -79.734377
Iteration 3: log likelihood = -79.150294
Iteration 4: log likelihood = -79.148124
Iteration 5: log likelihood = -79.148124
Logistic regression Number of obs = 143
LR chi2(2) = 39.38
Prob > chi2 = 0.0000
Log likelihood = -79.148124 Pseudo R2 = 0.1992
------------------------------------------------------------------------------
change | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_hat | 1.018599 .2037787 5.00 0.000 .6192005 1.417998
_hatsq | -.1179535 .129042 -0.91 0.361 -.3708711 .1349641
_cons | .1115148 .2294243 0.49 0.627 -.3381485 .5611782
------------------------------------------------------------------------------
Note: 1 failure and 0 successes completely determined.Code:
logit change firmsize profitability leverage age capitalintensity CAPEX KZindex elektricityge
> nerator Carbonleakage i.twodigitsNACE WestFlanders Hainaut Antwerp Brussels FlemishBrabant Li
> mbourg Liege Namur WalloonBrabant Luxembourg SME publicfirm, vce(robust)
note: 16.twodigitsNACE != 0 predicts success perfectly
16.twodigitsNACE dropped and 2 obs not used
note: 21.twodigitsNACE != 0 predicts success perfectly
21.twodigitsNACE dropped and 3 obs not used
note: 25.twodigitsNACE != 0 predicts success perfectly
25.twodigitsNACE dropped and 2 obs not used
note: 28.twodigitsNACE != 0 predicts success perfectly
28.twodigitsNACE dropped and 1 obs not used
note: 30.twodigitsNACE != 0 predicts success perfectly
30.twodigitsNACE dropped and 1 obs not used
note: 42.twodigitsNACE != 0 predicts success perfectly
42.twodigitsNACE dropped and 5 obs not used
note: 47.twodigitsNACE != 0 predicts failure perfectly
47.twodigitsNACE dropped and 1 obs not used
note: 49.twodigitsNACE != 0 predicts success perfectly
49.twodigitsNACE dropped and 2 obs not used
note: 61.twodigitsNACE != 0 predicts failure perfectly
61.twodigitsNACE dropped and 1 obs not used
note: 63.twodigitsNACE != 0 predicts failure perfectly
63.twodigitsNACE dropped and 2 obs not used
note: 70.twodigitsNACE != 0 predicts failure perfectly
70.twodigitsNACE dropped and 1 obs not used
note: 72.twodigitsNACE != 0 predicts success perfectly
72.twodigitsNACE dropped and 1 obs not used
note: 81.twodigitsNACE != 0 predicts success perfectly
81.twodigitsNACE dropped and 1 obs not used
note: Namur != 0 predicts failure perfectly
Namur dropped and 1 obs not used
Iteration 0: log pseudolikelihood = -98.836643
Iteration 1: log pseudolikelihood = -79.803416
Iteration 2: log pseudolikelihood = -79.587861
Iteration 3: log pseudolikelihood = -79.566988
Iteration 4: log pseudolikelihood = -79.565487
Iteration 5: log pseudolikelihood = -79.565147
Iteration 6: log pseudolikelihood = -79.565092
Iteration 7: log pseudolikelihood = -79.565086
Logistic regression Number of obs = 143
Wald chi2(33) = 210.53
Prob > chi2 = 0.0000
Log pseudolikelihood = -79.565086 Pseudo R2 = 0.1950
--------------------------------------------------------------------------------------
| Robust
change | Coef. Std. Err. z P>|z| [95% Conf. Interval]
---------------------+----------------------------------------------------------------
firmsize | .5369456 .2694897 1.99 0.046 .0087555 1.065136
profitability | -4.871327 6.919936 -0.70 0.481 -18.43415 8.691497
leverage | -2.234129 .9677091 -2.31 0.021 -4.130804 -.3374542
age | .0072455 .0099999 0.72 0.469 -.0123539 .0268449
capitalintensity | -1.686705 1.266685 -1.33 0.183 -4.169363 .7959533
CAPEX | 1.105687 3.557574 0.31 0.756 -5.867029 8.078404
KZindex | .2451881 .164491 1.49 0.136 -.0772083 .5675845
elektricitygenerator | 12.2858 2.103114 5.84 0.000 8.163768 16.40782
Carbonleakage | -.625777 .5255647 -1.19 0.234 -1.655865 .4043108
|
twodigitsNACE |
10 | -.2720088 1.375848 -0.20 0.843 -2.968622 2.424604
11 | 2.094524 1.867219 1.12 0.262 -1.565159 5.754207
13 | .2188543 1.611141 0.14 0.892 -2.938923 3.376632
16 | 0 (empty)
17 | -.4607484 1.592633 -0.29 0.772 -3.582253 2.660756
19 | .2352189 1.617443 0.15 0.884 -2.934911 3.405349
20 | .2973133 1.314841 0.23 0.821 -2.279728 2.874354
21 | 0 (empty)
22 | -.5714206 1.6593 -0.34 0.731 -3.823588 2.680747
23 | 1.222801 1.255445 0.97 0.330 -1.237825 3.683427
24 | .8195506 1.395696 0.59 0.557 -1.915963 3.555064
25 | 0 (empty)
28 | 0 (empty)
29 | .0334485 1.657259 0.02 0.984 -3.214719 3.281616
30 | 0 (empty)
35 | -12.92753 2.057606 -6.28 0.000 -16.96036 -8.894691
42 | 0 (empty)
46 | .638123 1.400915 0.46 0.649 -2.107621 3.383867
47 | 0 (empty)
49 | 0 (empty)
52 | 1.324916 1.606922 0.82 0.410 -1.824592 4.474425
61 | 0 (empty)
63 | 0 (empty)
70 | 0 (empty)
72 | 0 (empty)
81 | 0 (empty)
|
WestFlanders | -.0334296 .8781885 -0.04 0.970 -1.754647 1.687788
Hainaut | -1.990829 .7991805 -2.49 0.013 -3.557194 -.424464
Antwerp | -1.622771 .6982627 -2.32 0.020 -2.99134 -.2542008
Brussels | -2.43887 .8798601 -2.77 0.006 -4.163364 -.7143759
FlemishBrabant | .5337495 1.117561 0.48 0.633 -1.656631 2.724129
Limbourg | -.6921454 .8206851 -0.84 0.399 -2.300659 .9163678
Liege | -1.329429 1.024602 -1.30 0.194 -3.337612 .678754
Namur | 0 (omitted)
WalloonBrabant | -2.447982 1.391213 -1.76 0.078 -5.174709 .2787442
Luxembourg | -1.161801 1.30263 -0.89 0.372 -3.714908 1.391307
SME | 1.970419 .7692936 2.56 0.010 .4626317 3.478207
publicfirm | .1173165 2.651193 0.04 0.965 -5.078926 5.313559
_cons | -8.413368 5.235232 -1.61 0.108 -18.67423 1.847498
--------------------------------------------------------------------------------------
.Kind regards,
Timea De Wispelaere
0 Response to robust/bootstrap logistic regression
Post a Comment