I would also like to control for state and month FE. This is where it gets tricky. So, were I to exclude state and month FE, I would set up my logit model as follows. The first four independent variables are interactions of the binary policy variables and the time-period variable. The output is normal; no dropped variables (w/ exception of one very small occupational category)
Code:
logit reemp3 i.rq##i.prepostperiod_ma i.hidh##i.prepostperiod_ma i.lorh##i.prepostperiod_ma i.lorech##i.prepostperiod_ma b3.age_group b1.race_wbho b4.edu4 i.woman##i.marstdum1##i.ownkidd_18 b1.ind_nilf b1.uh_occmaj_b2 i.sampjl b1.durg ur_sa ur2_sa ur3_sa iur iur2 iur3 initrate initrate2 initrate3 empgrowth emp2 emp3 incrate_jhu stringd i.cutoff3n if sampall==1 & age>=18 & age<65 [pw=wtfinl], vce(cluster statefip) or
Code:
logit reemp3 i.rq i.hidh i.lorh i.lorech i.rq#i.prepostperiod_ma i.hidh#i.prepostperiod_ma i.lorh#i.prepostperiod_ma i.lorech#i.prepostperiod_ma b3.age_group b1.race_wbho b4.edu4 i.woman##i.marstdum1##i.ownkidd_18 b1.ind_nilf b1.uh_occmaj_b2 i.sampjl b1.durg ur_sa ur2_sa ur3_sa iur iur2 iur3 initrate initrate2 initrate3 empgrowth emp2 emp3 incrate_jhu stringd i.cutoff3n i.statefip i.ymd9 i.ymd10 i.ymd11 i.ymd12 i.ymd13 i.ymd14 i.ymd15 i.ymd16 i.ymd17 i.ymd18 i.ymd19 i.ymd20 if sampall==1 & age>=18 & age<65 [pw=wtfinl], vce(cluster statefip) or
Code:
. logit reemp3 i.rq i.hidh i.lorh i.lorech i.rq#i.prepostperiod_ma i.hidh#i.prepostperiod_ma i.lorh#i.prepostperiod_ma i.lorech#i.prepostperiod_
> ma b3.age_group b1.race_wbho b4.edu4 i.woman##i.marstdum1##i.ownkidd_18 b1.ind_nilf b1.uh_occmaj_b2 i.sampjl b1.durg ur_sa ur2_sa ur3_sa iur i
> ur2 iur3 initrate initrate2 initrate3 empgrowth emp2 emp3 incrate_jhu stringd i.cutoff3n i.statefip i.ymd9 i.ymd10 i.ymd11 i.ymd12 i.ymd13 i.y
> md14 i.ymd15 i.ymd16 i.ymd17 i.ymd18 i.ymd19 i.ymd20 if sampall==1 & age>=18 & age<65 [pw=wtfinl], vce(cluster statefip) or
note: 1.rq#1.prepostperiod_ma omitted because of collinearity.
note: 1.hidh#1.prepostperiod_ma omitted because of collinearity.
note: 1.lorh#1.prepostperiod_ma omitted because of collinearity.
note: 1.lorech#1.prepostperiod_ma omitted because of collinearity.
note: 11.uh_occmaj_b2 omitted because of collinearity.
note: 51.statefip omitted because of collinearity.
note: 53.statefip omitted because of collinearity.
note: 54.statefip omitted because of collinearity.
note: 55.statefip omitted because of collinearity.
note: 56.statefip omitted because of collinearity.
note: 1.ymd20 omitted because of collinearity.
Iteration 0: log pseudolikelihood = -22437923
Iteration 1: log pseudolikelihood = -21171234
Iteration 2: log pseudolikelihood = -21126355
Iteration 3: log pseudolikelihood = -21126157
Iteration 4: log pseudolikelihood = -21126157
Logistic regression Number of obs = 10,804
Wald chi2(43) = .
Prob > chi2 = .
Log pseudolikelihood = -21126157 Pseudo R2 = 0.0585
(Std. err. adjusted for 45 clusters in statefip)
--------------------------------------------------------------------------------------------------------------------
| Robust
reemp3 | Odds ratio std. err. z P>|z| [95% conf. interval]
---------------------------------------------------+----------------------------------------------------------------
1.rq | .8759639 .2052028 -0.57 0.572 .5534574 1.386399
1.hidh | .6605953 .1239371 -2.21 0.027 .4573386 .954186
1.lorh | 1.063168 .082126 0.79 0.428 .9137966 1.236956
1.lorech | .8994186 .1450101 -0.66 0.511 .6557309 1.233667
|
rq#prepostperiod_ma |
0 1 | 1.394841 .1676659 2.77 0.006 1.102063 1.765399
1 1 | 1 (omitted)
|
hidh#prepostperiod_ma |
0 1 | .9149866 .1208369 -0.67 0.501 .70632 1.185299
1 1 | 1 (omitted)
|
lorh#prepostperiod_ma |
0 1 | .9691477 .0887178 -0.34 0.732 .8099705 1.159607
1 1 | 1 (omitted)
|
lorech#prepostperiod_ma |
0 1 | .800244 .1520383 -1.17 0.241 .5514457 1.161294
1 1 | 1 (omitted)
|
age_group |
18-24 | .9053318 .0962845 -0.94 0.350 .7349879 1.115155
25-34 | .9163359 .0900103 -0.89 0.374 .7558619 1.11088
45-54 | .9557471 .0701927 -0.62 0.538 .8276151 1.103717
55-64 | .8012826 .0686457 -2.59 0.010 .6774286 .9477808
|
race_wbho |
2 black nh | .6752778 .0780661 -3.40 0.001 .5383672 .8470058
3 hispanic/latino | .9729026 .1120415 -0.24 0.811 .7763244 1.219258
other nh | .6696745 .0697663 -3.85 0.000 .5459919 .8213749
|
edu4 |
1 Less than HS | .9195733 .1157851 -0.67 0.505 .7184723 1.176963
2 HS or GED | .9592269 .0896774 -0.45 0.656 .7986255 1.152125
3 Some college or Associate's' | 1.025496 .0825591 0.31 0.754 .8758033 1.200774
|
1.woman | 1.057105 .0891657 0.66 0.510 .8960246 1.247142
1.marstdum1 | 1.42355 .1632607 3.08 0.002 1.136978 1.782351
|
woman#marstdum1 |
1 1 | .7660572 .1336128 -1.53 0.127 .5442492 1.078263
|
ownkidd_18 |
1: Own children, <18, in HH | .7174053 .1256241 -1.90 0.058 .5089927 1.011155
|
woman#ownkidd_18 |
1#1: Own children, <18, in HH | 1.387122 .2865239 1.58 0.113 .9253168 2.079404
|
marstdum1#ownkidd_18 |
1#1: Own children, <18, in HH | 1.284939 .3104339 1.04 0.299 .8002715 2.063137
|
woman#marstdum1#ownkidd_18 |
1#1#1: Own children, <18, in HH | .7495964 .2058639 -1.05 0.294 .437582 1.28409
|
ind_nilf |
2 | .294526 .1242367 -2.90 0.004 .1288459 .6732507
3 | .8449218 .3897243 -0.37 0.715 .3421336 2.086591
4 | .7630144 .3462639 -0.60 0.551 .313508 1.857021
5 | .7426147 .284665 -0.78 0.438 .3503275 1.574174
6 | .6189058 .2726112 -1.09 0.276 .2610318 1.467424
7 | .641471 .2559315 -1.11 0.266 .2934729 1.402122
8 | .9330527 .3859499 -0.17 0.867 .4147758 2.098935
9 | .625827 .2571176 -1.14 0.254 .2797324 1.400122
10 | .8877231 .3144192 -0.34 0.737 .443398 1.777302
11 | .5518519 .2135944 -1.54 0.125 .258443 1.178366
12 | .7979687 .3611106 -0.50 0.618 .3286891 1.937253
13 | .8836339 .3802865 -0.29 0.774 .3801401 2.054003
14 | .4842508 .4989079 -0.70 0.482 .0642843 3.64784
|
uh_occmaj_b2 |
professional and related occupations | 1.149595 .0924962 1.73 0.083 .9818773 1.34596
service occupations | 1.245314 .1368721 2.00 0.046 1.003976 1.544664
sales and related occupations | 1.144675 .1539027 1.00 0.315 .8795027 1.489797
office and administrative support occupations | 1.026844 .0943187 0.29 0.773 .8576676 1.229391
farming, fishing, and forestry occupations | .8031993 .3342352 -0.53 0.598 .3553145 1.815657
construction and extraction occupations | 1.057206 .1507861 0.39 0.697 .7993837 1.398183
installation, maintenance, and repair occupations | 1.174744 .2552418 0.74 0.459 .7673593 1.798405
production occupations | 1.212174 .1442737 1.62 0.106 .9599628 1.530649
transportation and material moving occupations | 1.143632 .1307977 1.17 0.241 .9139742 1.430996
armed forces | 1 (omitted)
|
1.sampjl | 1.459405 .1067366 5.17 0.000 1.264507 1.684341
|
durg |
5-8 weeks | .7148107 .0386471 -6.21 0.000 .6429391 .7947166
9-12 weeks | .6361483 .0478035 -6.02 0.000 .5490282 .7370927
13-16 weeks | .5037342 .0474485 -7.28 0.000 .4188164 .6058696
17-20 weeks | .4016522 .0605108 -6.05 0.000 .2989597 .5396197
21-26 weeks | .4995176 .0538834 -6.43 0.000 .4043253 .6171215
27-32 weeks | .3555218 .0883652 -4.16 0.000 .2184237 .5786726
33-38 weeks | .4384225 .0661046 -5.47 0.000 .3262497 .589163
39-44 weeks | .4156303 .1025017 -3.56 0.000 .2563221 .673951
45-50 weeks | .3319299 .1188662 -3.08 0.002 .1645226 .6696797
51-52 weeks | .3449078 .0759586 -4.83 0.000 .2239979 .5310826
>52 weeks | .2238139 .0389275 -8.61 0.000 .159162 .3147275
|
ur_sa | 3.30e-09 4.02e-08 -1.60 0.109 1.43e-19 76.45255
ur2_sa | 8.03e+56 6.18e+58 1.70 0.088 2.94e-09 2.2e+122
ur3_sa | 4.5e-135 6.8e-133 -2.02 0.043 3.6e-265 .0000561
iur | .0001876 .0021134 -0.76 0.446 4.85e-14 725261
iur2 | 6.00e+23 5.48e+25 0.60 0.549 1.12e-54 3.2e+101
iur3 | 1.91e-65 4.23e-63 -0.67 0.500 2.1e-253 1.8e+123
initrate | 2.58e-21 8.56e-20 -1.43 0.152 1.63e-49 4.10e+07
initrate2 | . . 1.59 0.112 5.2e-191 .
initrate3 | 0 0 -1.26 0.209 0 .
empgrowth | 1.026503 .0358656 0.75 0.454 .9585611 1.099262
emp2 | 1.002132 .0018383 1.16 0.246 .9985354 1.005741
emp3 | 1.000203 .0001074 1.89 0.059 .9999926 1.000414
incrate_jhu | .9996615 .0001283 -2.64 0.008 .9994101 .9999129
stringd | 1.009898 .0055966 1.78 0.076 .9989877 1.020927
1.cutoff3n | 1.204709 .1094021 2.05 0.040 1.008283 1.4394
|
statefip |
5 | .5474853 .1628406 -2.03 0.043 .3056303 .9807277
6 | .7088491 .0787637 -3.10 0.002 .5701283 .8813227
8 | .6208228 .1714694 -1.73 0.084 .3613003 1.066761
9 | .7189693 .1354696 -1.75 0.080 .4969644 1.040149
10 | .9189624 .1377839 -0.56 0.573 .6849731 1.232883
11 | .649819 .1030442 -2.72 0.007 .4762273 .8866874
13 | .9209017 .0895944 -0.85 0.397 .7610269 1.114363
15 | .9149104 .104705 -0.78 0.437 .7310794 1.144966
16 | 1.447961 .1215811 4.41 0.000 1.228241 1.706985
17 | .6666431 .1042136 -2.59 0.009 .4907141 .9056456
19 | 1.151438 .1322751 1.23 0.220 .919297 1.442198
20 | 1.228915 .1738412 1.46 0.145 .9313474 1.621556
21 | 1.139818 .1656932 0.90 0.368 .8572311 1.51556
23 | .6316927 .1850776 -1.57 0.117 .3557255 1.121752
24 | .6147948 .0446059 -6.70 0.000 .5333005 .7087422
25 | .6545012 .1038723 -2.67 0.008 .4795357 .8933056
26 | 2.299786 .637242 3.01 0.003 1.336073 3.958629
27 | .8193352 .0659604 -2.48 0.013 .6997386 .9593727
28 | 1.669741 .3574309 2.39 0.017 1.097584 2.540156
29 | 1.148785 .1399677 1.14 0.255 .9047499 1.458643
30 | 1.664098 .2325977 3.64 0.000 1.265328 2.188542
31 | 1.656605 .1573445 5.31 0.000 1.375219 1.995566
32 | 1.340935 .1634444 2.41 0.016 1.055981 1.702783
33 | .6338442 .1345354 -2.15 0.032 .418131 .9608436
34 | .8104846 .1040483 -1.64 0.102 .6301868 1.042366
35 | 1.050071 .1562071 0.33 0.743 .7845039 1.405537
36 | .7467416 .1153147 -1.89 0.059 .5517273 1.010686
37 | 1.003764 .1816936 0.02 0.983 .7039694 1.431231
38 | .735203 .0910294 -2.48 0.013 .576787 .9371285
40 | 1.012734 .1135587 0.11 0.910 .8129227 1.261657
41 | 1.082134 .2020508 0.42 0.672 .7504963 1.560318
42 | .8352292 .1290433 -1.17 0.244 .6170135 1.13062
44 | .8123845 .1488013 -1.13 0.257 .5673491 1.16325
45 | .7801361 .1288627 -1.50 0.133 .5643768 1.078379
46 | 1.158982 .111123 1.54 0.124 .9604249 1.398587
47 | 1.244327 .1920411 1.42 0.157 .9195294 1.683849
48 | 1.174027 .1647468 1.14 0.253 .8917274 1.545695
49 | .7521419 .1802977 -1.19 0.235 .4701715 1.203215
50 | 1.2256 .1100059 2.27 0.023 1.027893 1.461335
51 | 1 (omitted)
53 | 1 (omitted)
54 | 1 (omitted)
55 | 1 (omitted)
56 | 1 (omitted)
|
1.ymd9 | .5254365 .2632841 -1.28 0.199 .1967899 1.402936
1.ymd10 | .4938725 .2484605 -1.40 0.161 .1842413 1.323862
1.ymd11 | .4226503 .2169267 -1.68 0.093 .1545607 1.155748
1.ymd12 | .5955043 .2787812 -1.11 0.268 .2379035 1.490627
1.ymd13 | .5502248 .2544128 -1.29 0.196 .2223115 1.361816
1.ymd14 | .3811032 .1696496 -2.17 0.030 .1592674 .9119235
1.ymd15 | .6239076 .3885574 -0.76 0.449 .1840809 2.114618
1.ymd16 | .8106054 .1555722 -1.09 0.274 .5564756 1.180791
1.ymd17 | 1.066924 .1583095 0.44 0.662 .7976873 1.427034
1.ymd18 | 1.035925 .1403874 0.26 0.795 .7942812 1.351083
1.ymd19 | 1.034614 .1145833 0.31 0.759 .8327376 1.285431
1.ymd20 | 1 (omitted)
_cons | 2.566342 2.396611 1.01 0.313 .4115374 16.00368
--------------------------------------------------------------------------------------------------------------------
Note: _cons estimates baseline odds.
0 Response to Dealing w/ multicollinearity in logit model w/ FE (state, month)
Post a Comment