Hi all,
I am estimating a -heckprob- model. I got a rho which is significant. Then I
drop one variable which is not at all significant (it is zero) from the main
equation and estimate the model again. In this case I get a rho which is not
significant. I compare the standard errors and z-values in the two model
estimations and they are almost the same. So if this is the case how come
the rho becomes insignificant?
Thank you in advance for your time,
Irena
The following are the estimated models
. heckprob famtake child lnwwage howner age16_24 age25_34 age45_54 age55_64
NH_black NH_other Hispanic ed_hcg ed_co_at ed_co_g hialtimp hsvgex hsfapo
iphmo ipopt1 ilongtc impdummy_costall fpremnew1 sp1bp uncov [pw=pweight] if
ft_pt == 1 & classw == 1 & cont == 0 & couples == 1 & sftime ~= . & esex ==
2 , sel(off_hi = jobten f24_99 f100 location uncov occ_pro occ_tec occ_ser
occ_cra ind_con ind_man ind_tcp ind_tra ind_ser child lnwwage howner
age16_24 age25_34 age45_54 age55_64 NH_black NH_other Hispanic ed_hcg
ed_co_at ed_co_g hsvgex hsfapo midwest south west sp1bp msa) nolog;
Probit model with sample selection Number of obs = 1792
Censored obs = 431
Uncensored obs = 1361
Wald chi2(23) = 126.46
Log likelihood = -.0370687 Prob > chi2 = 0.0000
----------------------------------------------------------------------------
| Robust
| Coef. Std. Err. z P>|z| [95% Conf.
Interval]
-------------+----------------------------------------------------------
famtake |
child | -.1871534 .0700846 -2.67 0.008 -.3245167 -.0497902
lnwwage | .1571396 .0481592 3.26 0.001 .0627493 .25153
howner | .0062493 .0901974 0.07 0.945 -.1705344 .183033
age16_24 | -.066047 .156915 -0.42 0.674 -.3735946 .2415007
age25_34 | .1928288 .0795813 2.42 0.015 .0368523 .3488054
age45_54 | -.0154522 .0843621 -0.18 0.855 -.1807989 .1498946
age55_64 | .336145 .1864049 1.80 0.071 -.0292019 .701492
NH_black | .0651967 .1249824 0.52 0.602 -.1797643 .3101577
NH_other | -.1982407 .1612519 -1.23 0.219 -.5142887 .1178072
Hispanic | .1754764 .1273933 1.38 0.168 -.0742099 .4251627
ed_hcg | .0787785 .1247059 0.63 0.528 -.1656406 .3231975
ed_co_at | -.0412415 .127006 -0.32 0.745 -.2901686 .2076856
ed_co_g | .0640099 .1339407 0.48 0.633 -.198509 .3265289
hialtimp | -.3094602 .0951492 -3.25 0.001 -.4959491 -.1229713
hsvgex | -.077471 .0770738 -1.01 0.315 -.2285328 .0735908
hsfapo | .2486082 .149024 1.67 0.095 -.0434735 .5406899
iphmo | -.0237626 .0639877 -0.37 0.710 -.1491761 .1016509
ipopt1 | .0190663 .0649598 0.29 0.769 -.1082526 .1463852
ilongtc | .1754265 .0817656 2.15 0.032 .0151689 .3356841
impdummy | 1.020238 .3960393 2.58 0.010 .2440152 1.796461
fpremnew1 | -.0307759 .0150598 -2.04 0.041 -.0602925 -.0012592
sp1bp | -.5143257 .0965553 -5.33 0.000 -.7035705 -.3250808
uncov | .4932095 .1520762 3.24 0.001 .1951456 .7912734
_cons | .1053266 .324377 0.32 0.745 -.5304406 .7410938
--------+---------------------------------------------------------------
off_hi |
jobten | .009091 .0031228 2.91 0.004 .0029705 .0152116
f24_99 | .320946 .0716723 4.48 0.000 .1804708 .4614212
f100 | .5003315 .0594504 8.42 0.000 .3838109 .616852
location | .0783068 .0484032 1.62 0.106 -.0165617 .1731753
uncov | -.1266639 .0700653 -1.81 0.071 -.2639893 .0106616
occ_pro | -.0688214 .0890689 -0.77 0.440 -.2433932 .1057504
occ_tec | -.0279554 .0781832 -0.36 0.721 -.1811916 .1252808
occ_ser | -.3939632 .0957835 -4.11 0.000 -.5816954 -.2062309
occ_cra | .2157834 .1065291 2.03 0.043 .0069901 .4245766
ind_con | .7461699 .3043941 2.45 0.014 .1495685 1.342771
ind_man | .6786677 .2427632 2.80 0.005 .2028606 1.154475
ind_tcp | .8918545 .2551036 3.50 0.000 .3918606 1.391848
ind_tra | .6448755 .247033 2.61 0.009 .1606998 1.129051
ind_ser | .6479221 .2483057 2.61 0.009 .161252 1.134592
child | -.1936227 .0412623 -4.69 0.000 -.2744953 -.11275
lnwwage | .2433131 .0344783 7.06 0.000 .1757368 .3108894
howner | -.1156325 .0524034 -2.21 0.027 -.2183413 -.0129236
age16_24 | -.2455659 .0950341 -2.58 0.010 -.4318293 -.0593025
age25_34 | .0341946 .0466448 0.73 0.464 -.0572275 .1256167
age45_54 | -.1277697 .0515117 -2.48 0.013 -.2287307 -.0268087
age55_64 | -.5105788 .0974634 -5.24 0.000 -.7016035 -.319554
NH_black | -.235221 .0698127 -3.37 0.001 -.3720513 -.0983907
NH_other | .0749529 .0905079 0.83 0.408 -.1024393 .2523451
Hispanic | -.1647334 .0785995 -2.10 0.036 -.3187856 -.0106812
ed_hcg | -.0401284 .0875603 -0.46 0.647 -.2117434 .1314865
ed_co_at | -.0361566 .0913575 -0.40 0.692 -.215214 .1429008
ed_co_g | .0242729 .1031185 0.24 0.814 -.1778356 .2263815
hsvgex | .1058638 .0427342 2.48 0.013 .0221064 .1896212
hsfapo | .0892144 .0859086 1.04 0.299 -.0791633 .2575922
midwest | .0078949 .0588783 0.13 0.893 -.1075044 .1232942
south | .2603019 .0572391 4.55 0.000 .1481154 .3724885
west | -.0515523 .0662672 -0.78 0.437 -.1814336 .0783289
sp1bp | -.0333783 .05321 -0.63 0.530 -.137668 .0709113
msa | -.0720553 .0425368 -1.69 0.090 -.1554259 .0113153
_cons | -1.625491 .3132208 -5.19 0.000 -2.239393 -1.01159
---------+--------------------------------------------------------------
/athrho | -6.977666 2.198465 -3.17 0.002 -11.28658 -2.668754
---------+--------------------------------------------------------------
rho | -.9999983 7.65e-06 -1 -.9904303
----------------------------------------------------------------------------
Wald test of indep. eqns. (rho = 0): chi2(1) = 10.07 Prob > chi2 = 0.0015
----------------------------------------------------------------------------
Here I do drop ipopt1 variable:
. heckprob famtake child lnwwage howner age16_24 age25_34 age45_54 age55_64
NH_black NH_other Hispanic ed_hcg ed_co_at ed_co_g hialtimp hsvgex hsfapo
iphmo ilongtc impdummy_costall fpremnew1 sp1bp uncov [pw=pweight] if ft_pt
== 1 & classw == 1 & cont == 0 & couples == 1 & sftime ~= . & esex == 2 ,
sel(off_hi = jobten f24_99 f100 location uncov occ_pro occ_tec occ_ser
occ_cra ind_con ind_man ind_tcp ind_tra ind_ser child lnwwage howner
age16_24 age25_34 age45_54 age55_64 NH_black NH_other Hispanic ed_hcg
ed_co_at ed_co_g hsvgex hsfapo midwest south west sp1bp msa) nolog;
Probit model with sample selection Number of obs = 1792
Censored obs = 431
Uncensored obs = 1361
Wald chi2(22) = 126.74
Log likelihood = -.0370698 Prob > chi2 = 0.0000
----------------------------------------------------------------------------
| Robust
| Coef. Std. Err. z P>|z| [95% Conf.
Interval]
-------------+----------------------------------------------------------
famtake |
child | -.18494 .069155 -2.67 0.007 -.3204814 -.0493986
lnwwage | .1552191 .0536334 2.89 0.004 .0500996 .2603386
howner | .0075571 .0903043 0.08 0.933 -.1694361 .1845503
age16_24 | -.066341 .1572212 -0.42 0.673 -.3744889 .2418069
age25_34 | .193025 .0799033 2.42 0.016 .0364174 .3496327
age45_54 | -.0146542 .084174 -0.17 0.862 -.1796322 .1503238
age55_64 | .3360296 .1864588 1.80 0.072 -.029423 .7014821
NH_black | .0657508 .1250731 0.53 0.599 -.179388 .3108895
NH_other | -.1981662 .1610924 -1.23 0.219 -.5139016 .1175692
Hispanic | .175589 .1289784 1.36 0.173 -.0772039 .428382
ed_hcg | .0782575 .1249161 0.63 0.531 -.1665735 .3230884
ed_co_at | -.0427466 .1295397 -0.33 0.741 -.2966398 .2111466
ed_co_g | .0628189 .1357872 0.46 0.644 -.2033191 .3289569
hialtimp | -.3128602 .0945621 -3.31 0.001 -.4981985 -.127522
hsvgex | -.0766775 .0769418 -1.00 0.319 -.2274807 .0741256
hsfapo | .2493191 .1697997 1.47 0.142 -.0834823 .5821204
iphmo | -.025972 .0632996 -0.41 0.682 -.1500369 .0980928
ilongtc | .1751292 .0847913 2.07 0.039 .0089413 .3413172
impdummyl | 1.014092 .3949986 2.57 0.010 .2399089 1.788275
fpremnew1 | -.0306048 .0150421 -2.03 0.042 -.0600868 -.0011228
sp1bp | -.5150326 .0972594 -5.30 0.000 -.7056575 -.3244076
uncov | .4912066 .1519903 3.23 0.001 .193311 .7891022
_cons | .1281663 .3347733 0.38 0.702 -.5279774 .7843099
----------+-------------------------------------------------------------
off_hi |
jobten | .0090212 .003127 2.88 0.004 .0028923 .01515
f24_99 | .3202252 .0722729 4.43 0.000 .1785729 .4618775
f100 | .5004582 .0596404 8.39 0.000 .3835653 .6173512
location | .0784203 .048466 1.62 0.106 -.0165714 .173412
uncov | -.1261799 .0698181 -1.81 0.071 -.2630209 .0106611
occ_pro | -.068916 .0890796 -0.77 0.439 -.2435088 .1056767
occ_tec | -.0287224 .0782359 -0.37 0.714 -.182062 .1246171
occ_ser | -.3951517 .0962283 -4.11 0.000 -.5837558 -.2065476
occ_cra | .2168732 .106553 2.04 0.042 .0080331 .4257132
ind_con | .7490792 .3045025 2.46 0.014 .1522653 1.345893
ind_man | .6788584 .2427418 2.80 0.005 .2030931 1.154624
ind_tcp | .8931158 .2552474 3.50 0.000 .3928402 1.393391
ind_tra | .6469066 .2470374 2.62 0.009 .1627221 1.131091
ind_ser | .6481893 .2484943 2.61 0.009 .1611493 1.135229
child | -.1925664 .0411162 -4.68 0.000 -.2731526 -.1119801
lnwwage | .2420969 .0342239 7.07 0.000 .1750192 .3091746
howner | -.1153492 .0523461 -2.20 0.028 -.2179457 -.0127527
age16_24 | -.2465314 .0950661 -2.59 0.010 -.4328575 -.0602053
age25_34 | .0337855 .0465366 0.73 0.468 -.0574245 .1249955
age45_54 | -.127083 .0510015 -2.49 0.013 -.2270441 -.027122
age55_64 | -.5104681 .0970006 -5.26 0.000 -.7005859 -.3203504
NH_black | -.2346917 .0698471 -3.36 0.001 -.3715896 -.0977938
NH_other | .0731135 .0902959 0.81 0.418 -.1038632 .2500901
Hispanic | -.1658401 .0778087 -2.13 0.033 -.3183423 -.0133379
ed_hcg | -.0406197 .0876189 -0.46 0.643 -.2123496 .1311102
ed_co_at | -.0366932 .0917679 -0.40 0.689 -.2165549 .1431685
ed_co_g | .0238768 .1032489 0.23 0.817 -.1784873 .2262408
hsvgex | .1062537 .0427773 2.48 0.013 .0224117 .1900956
hsfapo | .0896835 .0836633 1.07 0.284 -.0742935 .2536604
midwest | .0083621 .0587307 0.14 0.887 -.106748 .1234723
south | .2611848 .0572153 4.56 0.000 .1490449 .3733247
west | -.0497923 .066289 -0.75 0.453 -.1797164 .0801318
sp1bp | -.0326143 .0530616 -0.61 0.539 -.1366132 .0713845
msa | -.0721723 .0426377 -1.69 0.091 -.1557407 .011396
_cons | -1.619899 .3133526 -5.17 0.000 -2.234059 -1.00574
----------+-------------------------------------------------------------
/athrho | -7.637805 6.169951 -1.24 0.216 -19.73069 4.455076
----------+-------------------------------------------------------------
rho | -.9999995 5.73e-06 -1 .99973
----------------------------------------------------------------------------
Wald test of indep. eqns. (rho = 0): chi2(1) = 1.53 Prob > chi2 = 0.2158
----------------------------------------------------------------------------
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
