I came across a lot of posts that faced a similar problem as mine. However, I still didn't find the exact answer I was looking for.
I am running a MNLogit regression:
These are a lot of variables, but in total I have 12,800 observations so it should be fine.
The "Note" that I receive is the following:
However, this warning message says that there are more observations than I have in my sample. I am assuming this is due to the fact that I use weights.
Without weights, there are ""only""" 534 observations completely determined.
In any case, it is clear that something is wrong. Ideally, I would like to examine those 534 observations that are completely determined. Is there a way to figure out which observations these are?
Thanks a lot in advance!
I am running a MNLogit regression:
Code:
mlogit croptype_dummy mean_DRO elevmean elevrange naturalness mean_HFO pdnsty SE005 SE425 SE025 irrigation_dummy0 ts1 ts1sq ts2 ts2sq ts3 ts3sq ts4 ts4sq ps1 ps1sq ps2 ps2sq ps3 ps3sq ps4 ps4sq rentedland gdpcap t_gravel t_silt t_sand t_ph_h2o lat lon subsidies1 [fweight=land], robust
The "Note" that I receive is the following:
Code:
Note: 174057 observations completely determined. Standard errors questionable.
Without weights, there are ""only""" 534 observations completely determined.
In any case, it is clear that something is wrong. Ideally, I would like to examine those 534 observations that are completely determined. Is there a way to figure out which observations these are?
Thanks a lot in advance!
Code:
Iteration 0: log pseudolikelihood = -720199.8
Iteration 1: log pseudolikelihood = -511838.18
Iteration 2: log pseudolikelihood = -399636.45
Iteration 3: log pseudolikelihood = -339029.07
Iteration 4: log pseudolikelihood = -286488.06
Iteration 5: log pseudolikelihood = -268131.02
Iteration 6: log pseudolikelihood = -258566.26
Iteration 7: log pseudolikelihood = -255407.41
Iteration 8: log pseudolikelihood = -254630.09
Iteration 9: log pseudolikelihood = -254592.04
Iteration 10: log pseudolikelihood = -254591.86
Iteration 11: log pseudolikelihood = -254591.86
Multinomial logistic regression Number of obs = 894,740
Wald chi2(105) = 202031.19
Prob > chi2 = 0.0000
Log pseudolikelihood = -254591.86 Pseudo R2 = 0.6465
----------------------------------------------------------------------------------
| Robust
croptype_dummy | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-----------------+----------------------------------------------------------------
2 |
mean_DRO | .0190114 .0018295 10.39 0.000 .0154256 .0225972
elevmean | -3.878067 .1058287 -36.64 0.000 -4.085488 -3.670647
elevrange | .150083 .0198067 7.58 0.000 .1112626 .1889033
naturalness | 1.993495 .0353167 56.45 0.000 1.924275 2.062714
mean_HFO | -.0142505 .002723 -5.23 0.000 -.0195874 -.0089135
pdnsty | .9211858 .0405922 22.69 0.000 .8416265 1.000745
SE005 | 34.32297 .1926833 178.13 0.000 33.94532 34.70062
SE425 | -.0077633 .0003862 -20.10 0.000 -.0085202 -.0070063
SE025 | -.0629448 .0003565 -176.57 0.000 -.0636435 -.0622461
irrigation_du~y0 | 1.424357 .0195388 72.90 0.000 1.386062 1.462652
ts1 | .4414631 .0389151 11.34 0.000 .3651909 .5177352
ts1sq | -.0203034 .0031534 -6.44 0.000 -.0264841 -.0141228
ts2 | -.9033541 .0910221 -9.92 0.000 -1.081754 -.7249541
ts2sq | -.0029859 .0045208 -0.66 0.509 -.0118465 .0058747
ts3 | -.8920773 .1350362 -6.61 0.000 -1.156743 -.6274113
ts3sq | .02114 .0035085 6.03 0.000 .0142635 .0280166
ts4 | -.3709673 .1376005 -2.70 0.007 -.6406593 -.1012752
ts4sq | .0268379 .0051513 5.21 0.000 .0167417 .0369342
ps1 | -1.940215 .034915 -55.57 0.000 -2.008647 -1.871783
ps1sq | .0946409 .0018467 51.25 0.000 .0910214 .0982603
ps2 | .5586826 .0746853 7.48 0.000 .4123022 .705063
ps2sq | -.0433847 .004466 -9.71 0.000 -.0521379 -.0346315
ps3 | -.4181258 .0423171 -9.88 0.000 -.5010657 -.3351858
ps3sq | .0328159 .0020096 16.33 0.000 .028877 .0367547
ps4 | .8771014 .0424684 20.65 0.000 .7938648 .960338
ps4sq | -.0539718 .0025632 -21.06 0.000 -.0589957 -.0489479
rentedland | -.8017062 .0263296 -30.45 0.000 -.8533113 -.7501012
gdpcap | .0106112 .0014131 7.51 0.000 .0078416 .0133808
t_gravel | -.2634573 .005928 -44.44 0.000 -.2750759 -.2518387
t_silt | .0970622 .0035782 27.13 0.000 .090049 .1040755
t_sand | .0287123 .0023261 12.34 0.000 .0241533 .0332713
t_ph_h2o | .0871085 .0270917 3.22 0.001 .0340098 .1402073
lat | -.3463195 .0115571 -29.97 0.000 -.3689711 -.323668
lon | .0037603 .0027734 1.36 0.175 -.0016755 .0091961
subsidies1 | .8798013 .025646 34.31 0.000 .8295362 .9300665
_cons | 23.4252 1.081689 21.66 0.000 21.30513 25.54527
-----------------+----------------------------------------------------------------
3 |
mean_DRO | .0325762 .0014126 23.06 0.000 .0298076 .0353449
elevmean | -2.287876 .1015219 -22.54 0.000 -2.486855 -2.088897
elevrange | -.8220184 .0166973 -49.23 0.000 -.8547446 -.7892923
naturalness | 1.451338 .0293013 49.53 0.000 1.393909 1.508768
mean_HFO | -.0550362 .002169 -25.37 0.000 -.0592873 -.0507852
pdnsty | .3972764 .0347746 11.42 0.000 .3291194 .4654335
SE005 | 34.18779 .1916273 178.41 0.000 33.81221 34.56337
SE425 | .0033927 .0001233 27.52 0.000 .0031511 .0036343
SE025 | -.0566438 .0003257 -173.93 0.000 -.0572822 -.0560055
irrigation_du~y0 | 1.627022 .0173354 93.86 0.000 1.593045 1.660999
ts1 | -.6976467 .0288475 -24.18 0.000 -.7541868 -.6411066
ts1sq | .06174 .0018054 34.20 0.000 .0582015 .0652785
ts2 | 1.071005 .0715424 14.97 0.000 .930784 1.211225
ts2sq | -.1327408 .0038482 -34.49 0.000 -.1402831 -.1251984
ts3 | -5.660249 .0978786 -57.83 0.000 -5.852088 -5.468411
ts3sq | .1625688 .0024185 67.22 0.000 .1578287 .167309
ts4 | 2.693855 .0992124 27.15 0.000 2.499402 2.888308
ts4sq | -.0785121 .0037492 -20.94 0.000 -.0858603 -.0711639
ps1 | -1.300247 .0306413 -42.43 0.000 -1.360303 -1.240191
ps1sq | .0783001 .0017888 43.77 0.000 .0747941 .0818061
ps2 | 2.334851 .0706095 33.07 0.000 2.196459 2.473243
ps2sq | -.1832527 .0047486 -38.59 0.000 -.1925597 -.1739457
ps3 | -.7822601 .0362545 -21.58 0.000 -.8533176 -.7112025
ps3sq | .0409215 .0019159 21.36 0.000 .0371665 .0446765
ps4 | -.5449703 .0334276 -16.30 0.000 -.6104873 -.4794533
ps4sq | .0353429 .0019948 17.72 0.000 .0314332 .0392526
rentedland | 1.001006 .020326 49.25 0.000 .9611675 1.040844
gdpcap | .0363596 .0011567 31.43 0.000 .0340925 .0386267
t_gravel | -.0695202 .0049207 -14.13 0.000 -.0791645 -.0598758
t_silt | .0917565 .003038 30.20 0.000 .0858022 .0977109
t_sand | .1015854 .0018848 53.90 0.000 .0978912 .1052796
t_ph_h2o | .6078819 .0212503 28.61 0.000 .566232 .6495319
lat | -.2060046 .0109167 -18.87 0.000 -.2274009 -.1846083
lon | -.0232922 .0021246 -10.96 0.000 -.0274563 -.0191281
subsidies1 | 1.349989 .0228129 59.18 0.000 1.305277 1.394702
_cons | 22.60974 1.034219 21.86 0.000 20.5827 24.63677
-----------------+----------------------------------------------------------------
5 |
mean_DRO | -.020908 .0019059 -10.97 0.000 -.0246436 -.0171725
elevmean | -1.388079 .0895233 -15.51 0.000 -1.563541 -1.212616
elevrange | -.3699196 .0134193 -27.57 0.000 -.3962209 -.3436183
naturalness | .2288814 .0294597 7.77 0.000 .1711415 .2866214
mean_HFO | -.1361742 .0022704 -59.98 0.000 -.1406242 -.1317242
pdnsty | -.6701923 .043301 -15.48 0.000 -.7550606 -.5853239
SE005 | 33.10582 .1930321 171.50 0.000 32.72748 33.48416
SE425 | -.0019956 .0001718 -11.62 0.000 -.0023324 -.0016589
SE025 | -.0535544 .0003305 -162.06 0.000 -.0542021 -.0529067
irrigation_du~y0 | -1.337171 .0164423 -81.33 0.000 -1.369397 -1.304945
ts1 | .9485747 .0289214 32.80 0.000 .8918899 1.005259
ts1sq | .1267372 .0021115 60.02 0.000 .1225988 .1308755
ts2 | 4.922767 .1033373 47.64 0.000 4.720229 5.125304
ts2sq | -.1778646 .005197 -34.22 0.000 -.1880505 -.1676787
ts3 | -2.451771 .0985995 -24.87 0.000 -2.645022 -2.258519
ts3sq | .1055752 .0023235 45.44 0.000 .1010212 .1101291
ts4 | -3.663043 .116101 -31.55 0.000 -3.890597 -3.435489
ts4sq | -.0629105 .0046134 -13.64 0.000 -.0719525 -.0538685
ps1 | -.5426029 .0287091 -18.90 0.000 -.5988716 -.4863341
ps1sq | .015288 .0015217 10.05 0.000 .0123056 .0182705
ps2 | -1.435326 .0609773 -23.54 0.000 -1.554839 -1.315813
ps2sq | .1374548 .0033183 41.42 0.000 .130951 .1439587
ps3 | .4154882 .0361699 11.49 0.000 .3445965 .4863799
ps3sq | -.0552709 .0019157 -28.85 0.000 -.0590256 -.0515163
ps4 | 1.369163 .0331446 41.31 0.000 1.3042 1.434125
ps4sq | -.0666496 .0018868 -35.32 0.000 -.0703477 -.0629516
rentedland | -.4546961 .0201763 -22.54 0.000 -.494241 -.4151512
gdpcap | .0298129 .0013739 21.70 0.000 .0271201 .0325057
t_gravel | -.032882 .0053869 -6.10 0.000 -.0434402 -.0223239
t_silt | .1168384 .0031453 37.15 0.000 .1106738 .1230029
t_sand | .0605906 .00205 29.56 0.000 .0565727 .0646085
t_ph_h2o | .3102023 .0206419 15.03 0.000 .269745 .3506596
lat | -.8060968 .0104488 -77.15 0.000 -.8265762 -.7856175
lon | .021756 .0025356 8.58 0.000 .0167864 .0267256
subsidies1 | -.6419808 .0362229 -17.72 0.000 -.7129764 -.5709851
_cons | 53.05945 1.140427 46.53 0.000 50.82426 55.29465
-----------------+----------------------------------------------------------------
7 | (base outcome)
----------------------------------------------------------------------------------
Note: 174057 observations completely determined. Standard errors questionable.
.
Comment