Using the PLFS data of various rounds, I am trying to know the determinants of elderly workers engagement in farm sector jobs, non-farm sector jobs and neither of the jobs. To do so, I am running a multinomial logit model. One of the variables is the predicted value of log of wage earning which I deduced from Heckman method. But when I run the mlogit model using the predicted values of log wage, some other variables got omitted due to collinearity. Can anyone please help me?
heckman logwageearning age agesquare sexdum2 , select(wpr = age agesquare sexdum2 castedum2 castedum3 castedum4 literac
> ydum2 literacydum3 literacydum4 literacydum5 formaledu_yrs logmpce logmpcesquare hh_size) twostep
Heckman selection model -- two-step estimates Number of obs = 260,015
(regression model with sample selection) Selected = 77,304
Nonselected = 182,711
Wald chi2(3) = 8558.97
Prob > chi2 = 0.0000
--------------------------------------------------------------------------------
| Coefficient Std. err. z P>|z| [95% conf. interval]
---------------+----------------------------------------------------------------
logwageearning |
age | -.1031633 .0444615 -2.32 0.020 -.1903063 -.0160203
agesquare | -.0004546 .000323 -1.41 0.159 -.0010876 .0001784
sexdum2 | -5.244314 .0750406 -69.89 0.000 -5.391391 -5.097237
_cons | 15.15177 1.524618 9.94 0.000 12.16357 18.13997
---------------+----------------------------------------------------------------
wpr |
age | -.1726896 .0083118 -20.78 0.000 -.1889805 -.1563987
agesquare | .0006723 .0000596 11.28 0.000 .0005555 .0007891
sexdum2 | -1.154555 .0063782 -181.02 0.000 -1.167056 -1.142054
castedum2 | -.0711323 .011402 -6.24 0.000 -.0934798 -.0487848
castedum3 | -.0550081 .0098303 -5.60 0.000 -.0742751 -.035741
castedum4 | -.1239962 .0102845 -12.06 0.000 -.1441535 -.1038389
literacydum2 | .101267 .0134382 7.54 0.000 .0749286 .1276055
literacydum3 | -.0604363 .022626 -2.67 0.008 -.1047823 -.0160902
literacydum4 | -.1524332 .0279429 -5.46 0.000 -.2072003 -.0976661
literacydum5 | -.2017025 .0332416 -6.07 0.000 -.2668548 -.1365502
formaledu_yrs | -.0196811 .0019925 -9.88 0.000 -.0235863 -.0157758
logmpce | .802094 .0813404 9.86 0.000 .6426697 .9615183
logmpcesquare | -.055815 .0052062 -10.72 0.000 -.0660189 -.0456111
hh_size | -.0254606 .0013346 -19.08 0.000 -.0280764 -.0228449
_cons | 5.887443 .4287958 13.73 0.000 5.047019 6.727868
---------------+----------------------------------------------------------------
/mills |
lambda | 2.249649 .094125 23.90 0.000 2.065167 2.43413
---------------+----------------------------------------------------------------
rho | 0.62136
sigma | 3.6205183
--------------------------------------------------------------------------------
. predict plogwage, xb
. mlogit sector_emp1 age agesquare logmpce logmpcesquare plogwage sexdum2 castedum1 castedum2 castedum3 religdum1 religdu
> m2 literacydum2 literacydum3 literacydum4 literacydum5, base(0)
note: sexdum2 omitted because of collinearity.
Iteration 0: log likelihood = -210400.55
Iteration 1: log likelihood = -175957.84
Iteration 2: log likelihood = -172847.83
Iteration 3: log likelihood = -172738.15
Iteration 4: log likelihood = -172737.04
Iteration 5: log likelihood = -172737.03
Multinomial logistic regression Number of obs = 260,167
LR chi2(28) = 75327.04
Prob > chi2 = 0.0000
Log likelihood = -172737.03 Pseudo R2 = 0.1790
-------------------------------------------------------------------------------
sector_emp1 | Coefficient Std. err. z P>|z| [95% conf. interval]
--------------+----------------------------------------------------------------
Neither | (base outcome)
--------------+----------------------------------------------------------------
Farm |
age | -.0971083 .0201615 -4.82 0.000 -.1366242 -.0575924
agesquare | .0001424 .000147 0.97 0.333 -.0001458 .0004306
logmpce | 3.390933 .1935149 17.52 0.000 3.011651 3.770215
logmpcesquare | -.2391877 .0126258 -18.94 0.000 -.2639339 -.2144415
plogwage | .3605951 .0024977 144.37 0.000 .3556996 .3654905
sexdum2 | 0 (omitted)
castedum1 | .4030476 .0208022 19.38 0.000 .362276 .4438192
castedum2 | -.065252 .0187967 -3.47 0.001 -.1020928 -.0284111
castedum3 | .0511478 .0142847 3.58 0.000 .0231503 .0791453
religdum1 | .1324626 .0201015 6.59 0.000 .0930644 .1718608
religdum2 | -.5077467 .0278075 -18.26 0.000 -.5622483 -.4532451
literacydum2 | -.2300108 .0135842 -16.93 0.000 -.2566354 -.2033862
literacydum3 | -.7649722 .0237101 -32.26 0.000 -.8114431 -.7185013
literacydum4 | -1.103101 .0340094 -32.44 0.000 -1.169759 -1.036444
literacydum5 | -1.692593 .0357945 -47.29 0.000 -1.762749 -1.622437
_cons | -8.708025 1.010458 -8.62 0.000 -10.68849 -6.727564
--------------+----------------------------------------------------------------
Non_Farm |
age | -.3595212 .0218435 -16.46 0.000 -.4023336 -.3167088
agesquare | .002031 .0001588 12.79 0.000 .0017197 .0023422
logmpce | 2.236612 .1924041 11.62 0.000 1.859507 2.613717
logmpcesquare | -.1289564 .0121408 -10.62 0.000 -.1527518 -.1051609
plogwage | .3979168 .0031529 126.21 0.000 .3917373 .4040964
sexdum2 | 0 (omitted)
castedum1 | -.4434678 .0306714 -14.46 0.000 -.5035826 -.3833529
castedum2 | .1882085 .0215112 8.75 0.000 .1460473 .2303697
castedum3 | .1589689 .0156608 10.15 0.000 .1282742 .1896636
religdum1 | .0511494 .0238179 2.15 0.032 .0044672 .0978316
religdum2 | .0823476 .030172 2.73 0.006 .0232116 .1414836
literacydum2 | .3850795 .0166694 23.10 0.000 .3524081 .4177508
literacydum3 | .0928973 .0246065 3.78 0.000 .0446694 .1411252
literacydum4 | -.074376 .0316077 -2.35 0.019 -.1363259 -.0124261
literacydum5 | -.1390339 .027936 -4.98 0.000 -.1937875 -.0842804
_cons | 1.568251 1.064079 1.47 0.141 -.5173047 3.653808
-------------------------------------------------------------------------------
heckman logwageearning age agesquare sexdum2 , select(wpr = age agesquare sexdum2 castedum2 castedum3 castedum4 literac
> ydum2 literacydum3 literacydum4 literacydum5 formaledu_yrs logmpce logmpcesquare hh_size) twostep
Heckman selection model -- two-step estimates Number of obs = 260,015
(regression model with sample selection) Selected = 77,304
Nonselected = 182,711
Wald chi2(3) = 8558.97
Prob > chi2 = 0.0000
--------------------------------------------------------------------------------
| Coefficient Std. err. z P>|z| [95% conf. interval]
---------------+----------------------------------------------------------------
logwageearning |
age | -.1031633 .0444615 -2.32 0.020 -.1903063 -.0160203
agesquare | -.0004546 .000323 -1.41 0.159 -.0010876 .0001784
sexdum2 | -5.244314 .0750406 -69.89 0.000 -5.391391 -5.097237
_cons | 15.15177 1.524618 9.94 0.000 12.16357 18.13997
---------------+----------------------------------------------------------------
wpr |
age | -.1726896 .0083118 -20.78 0.000 -.1889805 -.1563987
agesquare | .0006723 .0000596 11.28 0.000 .0005555 .0007891
sexdum2 | -1.154555 .0063782 -181.02 0.000 -1.167056 -1.142054
castedum2 | -.0711323 .011402 -6.24 0.000 -.0934798 -.0487848
castedum3 | -.0550081 .0098303 -5.60 0.000 -.0742751 -.035741
castedum4 | -.1239962 .0102845 -12.06 0.000 -.1441535 -.1038389
literacydum2 | .101267 .0134382 7.54 0.000 .0749286 .1276055
literacydum3 | -.0604363 .022626 -2.67 0.008 -.1047823 -.0160902
literacydum4 | -.1524332 .0279429 -5.46 0.000 -.2072003 -.0976661
literacydum5 | -.2017025 .0332416 -6.07 0.000 -.2668548 -.1365502
formaledu_yrs | -.0196811 .0019925 -9.88 0.000 -.0235863 -.0157758
logmpce | .802094 .0813404 9.86 0.000 .6426697 .9615183
logmpcesquare | -.055815 .0052062 -10.72 0.000 -.0660189 -.0456111
hh_size | -.0254606 .0013346 -19.08 0.000 -.0280764 -.0228449
_cons | 5.887443 .4287958 13.73 0.000 5.047019 6.727868
---------------+----------------------------------------------------------------
/mills |
lambda | 2.249649 .094125 23.90 0.000 2.065167 2.43413
---------------+----------------------------------------------------------------
rho | 0.62136
sigma | 3.6205183
--------------------------------------------------------------------------------
. predict plogwage, xb
. mlogit sector_emp1 age agesquare logmpce logmpcesquare plogwage sexdum2 castedum1 castedum2 castedum3 religdum1 religdu
> m2 literacydum2 literacydum3 literacydum4 literacydum5, base(0)
note: sexdum2 omitted because of collinearity.
Iteration 0: log likelihood = -210400.55
Iteration 1: log likelihood = -175957.84
Iteration 2: log likelihood = -172847.83
Iteration 3: log likelihood = -172738.15
Iteration 4: log likelihood = -172737.04
Iteration 5: log likelihood = -172737.03
Multinomial logistic regression Number of obs = 260,167
LR chi2(28) = 75327.04
Prob > chi2 = 0.0000
Log likelihood = -172737.03 Pseudo R2 = 0.1790
-------------------------------------------------------------------------------
sector_emp1 | Coefficient Std. err. z P>|z| [95% conf. interval]
--------------+----------------------------------------------------------------
Neither | (base outcome)
--------------+----------------------------------------------------------------
Farm |
age | -.0971083 .0201615 -4.82 0.000 -.1366242 -.0575924
agesquare | .0001424 .000147 0.97 0.333 -.0001458 .0004306
logmpce | 3.390933 .1935149 17.52 0.000 3.011651 3.770215
logmpcesquare | -.2391877 .0126258 -18.94 0.000 -.2639339 -.2144415
plogwage | .3605951 .0024977 144.37 0.000 .3556996 .3654905
sexdum2 | 0 (omitted)
castedum1 | .4030476 .0208022 19.38 0.000 .362276 .4438192
castedum2 | -.065252 .0187967 -3.47 0.001 -.1020928 -.0284111
castedum3 | .0511478 .0142847 3.58 0.000 .0231503 .0791453
religdum1 | .1324626 .0201015 6.59 0.000 .0930644 .1718608
religdum2 | -.5077467 .0278075 -18.26 0.000 -.5622483 -.4532451
literacydum2 | -.2300108 .0135842 -16.93 0.000 -.2566354 -.2033862
literacydum3 | -.7649722 .0237101 -32.26 0.000 -.8114431 -.7185013
literacydum4 | -1.103101 .0340094 -32.44 0.000 -1.169759 -1.036444
literacydum5 | -1.692593 .0357945 -47.29 0.000 -1.762749 -1.622437
_cons | -8.708025 1.010458 -8.62 0.000 -10.68849 -6.727564
--------------+----------------------------------------------------------------
Non_Farm |
age | -.3595212 .0218435 -16.46 0.000 -.4023336 -.3167088
agesquare | .002031 .0001588 12.79 0.000 .0017197 .0023422
logmpce | 2.236612 .1924041 11.62 0.000 1.859507 2.613717
logmpcesquare | -.1289564 .0121408 -10.62 0.000 -.1527518 -.1051609
plogwage | .3979168 .0031529 126.21 0.000 .3917373 .4040964
sexdum2 | 0 (omitted)
castedum1 | -.4434678 .0306714 -14.46 0.000 -.5035826 -.3833529
castedum2 | .1882085 .0215112 8.75 0.000 .1460473 .2303697
castedum3 | .1589689 .0156608 10.15 0.000 .1282742 .1896636
religdum1 | .0511494 .0238179 2.15 0.032 .0044672 .0978316
religdum2 | .0823476 .030172 2.73 0.006 .0232116 .1414836
literacydum2 | .3850795 .0166694 23.10 0.000 .3524081 .4177508
literacydum3 | .0928973 .0246065 3.78 0.000 .0446694 .1411252
literacydum4 | -.074376 .0316077 -2.35 0.019 -.1363259 -.0124261
literacydum5 | -.1390339 .027936 -4.98 0.000 -.1937875 -.0842804
_cons | 1.568251 1.064079 1.47 0.141 -.5173047 3.653808
-------------------------------------------------------------------------------
