use https://stats.idre.ucla.edu/stat/stata/examples/methods_matter/chapter10/dee, clear
(a) Univariate Statistics from Table 10.1 on page 207.
summarize register college, detail
Is respondent currently registered to vote?
-------------------------------------------------------------
Percentiles Smallest
1% 0 0
5% 0 0
10% 0 0 Obs 9227
25% 0 0 Sum of Wgt. 9227
50% 1 Mean .6708573
Largest Std. Dev. .4699274
75% 1 1
90% 1 1 Variance .2208317
95% 1 1 Skewness -.727204
99% 1 1 Kurtosis 1.528826
Attended junior, community or 4year college by
1984?
-------------------------------------------------------------
Percentiles Smallest
1% 0 0
5% 0 0
10% 0 0 Obs 9227
25% 0 0 Sum of Wgt. 9227
50% 1 Mean .5470901
Largest Std. Dev. .4978046
75% 1 1
90% 1 1 Variance .2478094
95% 1 1 Skewness -.1892012
99% 1 1 Kurtosis 1.035797
Test whether the means of register and college are significantly different from zero.
Note: this output does not appear in the table.
ttest register == 0
One-sample t test
------------------------------------------------------------------------------
Variable | Obs Mean Std. Err. Std. Dev. [95% Conf. Interval]
---------+--------------------------------------------------------------------
register | 9227 .6708573 .0048922 .4699274 .6612676 .680447
------------------------------------------------------------------------------
mean = mean(register) t = 137.1291
Ho: mean = 0 degrees of freedom = 9226
Ha: mean < 0 Ha: mean != 0 Ha: mean > 0
Pr(T < t) = 1.0000 Pr(|T| > |t|) = 0.0000 Pr(T > t) = 0.0000
ttest college == 0
One-sample t test
------------------------------------------------------------------------------
Variable | Obs Mean Std. Err. Std. Dev. [95% Conf. Interval]
---------+--------------------------------------------------------------------
college | 9227 .5470901 .0051824 .4978046 .5369315 .5572487
------------------------------------------------------------------------------
mean = mean(college) t = 105.5675
Ho: mean = 0 degrees of freedom = 9226
Ha: mean < 0 Ha: mean != 0 Ha: mean > 0
Pr(T < t) = 1.0000 Pr(|T| > |t|) = 0.0000 Pr(T > t) = 0.0000
Cross-tabulation of register and college, with
a chi-squared test. (Not shown in the text.)
tabulate register college, cchi2 chi2 expected
+--------------------+
| Key |
|--------------------|
| frequency |
| expected frequency |
| chi2 contribution |
+--------------------+
Is |
respondent | Attended junior,
currently | community or 4year
registered | college by 1984?
to vote? | 0 1 | Total
-----------+----------------------+----------
0 | 1,780 1,257 | 3,037
| 1,375.5 1,661.5 | 3,037.0
| 119.0 98.5 | 217.4
-----------+----------------------+----------
1 | 2,399 3,791 | 6,190
| 2,803.5 3,386.5 | 6,190.0
| 58.4 48.3 | 106.7
-----------+----------------------+----------
Total | 4,179 5,048 | 9,227
| 4,179.0 5,048.0 | 9,227.0
| 177.3 146.8 | 324.1
(b) Sample Bivariate Correlations and Covariances from Table 10.1 on page 207.
pwcorr register college, sig
| register college
-------------+------------------
register | 1.0000
|
|
college | 0.1874 1.0000
| 0.0000
|
correlate register college, covariance
(obs=9227)
| register college
-------------+------------------
register | .220832
college | .043845 .247809
(c) OLS Regression Analysis: Outcome=register from Table 10.1 on page 207.
regress register college
Source | SS df MS Number of obs = 9227
-------------+------------------------------ F( 1, 9225) = 335.86
Model | 71.5702829 1 71.5702829 Prob > F = 0.0000
Residual | 1965.82324 9225 .21309737 R-squared = 0.0351
-------------+------------------------------ Adj R-squared = 0.0350
Total | 2037.39352 9226 .220831728 Root MSE = .46162
------------------------------------------------------------------------------
register | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
college | .1769297 .0096544 18.33 0.000 .158005 .1958544
_cons | .5740608 .0071409 80.39 0.000 .560063 .5880585
------------------------------------------------------------------------------
(a) Univariate Statistics from Table 10.2 on page 219. Note that only the variable distance has
been added from panel a of Table 10.1.
summarize distance, detail
Miles from respondents HS to nearest 2yr college
-------------------------------------------------------------
Percentiles Smallest
1% 0 0
5% 1 0
10% 1 0 Obs 9227
25% 3 0 Sum of Wgt. 9227
50% 7 Mean 9.735992
Largest Std. Dev. 8.702286
75% 15 35
90% 25 35 Variance 75.72978
95% 30 35 Skewness 1.181081
99% 35 35 Kurtosis 3.599886
Test that the mean of distance is significantly different from zero. (Not shown in the
table.)
ttest distance == 0
One-sample t test
------------------------------------------------------------------------------
Variable | Obs Mean Std. Err. Std. Dev. [95% Conf. Interval]
---------+--------------------------------------------------------------------
distance | 9227 9.735992 .0905948 8.702286 9.558407 9.913578
------------------------------------------------------------------------------
mean = mean(distance) t = 107.4675
Ho: mean = 0 degrees of freedom = 9226
Ha: mean < 0 Ha: mean != 0 Ha: mean > 0
Pr(T < t) = 1.0000 Pr(|T| > |t|) = 0.0000 Pr(T > t) = 0.0000
(b) Sample Bivariate Correlations and Covariances from Table 10.2 on page 219.
pwcorr register college distance, sig
| register college distance
-------------+---------------------------
register | 1.0000
|
|
college | 0.1874 1.0000
| 0.0000
|
distance | -0.0335 -0.1114 1.0000
| 0.0013 0.0000
|
correlate register college distance, covariance
(obs=9227)
| register college distance
-------------+---------------------------
register | .220832
college | .043845 .247809
distance | -.136873 -.482472 75.7298
(c) Method-of-Moments IVE Estimate from Table 10.2 on page 219.
display -.136873/-.482472 .28369107
Table 10.3 on page 228.
ivreg register (college=distance), first
First-stage regressions
-----------------------
Source | SS df MS Number of obs = 9227
-------------+------------------------------ F( 1, 9225) = 115.86
Model | 28.3590322 1 28.3590322 Prob > F = 0.0000
Residual | 2257.93034 9225 .244762096 R-squared = 0.0124
-------------+------------------------------ Adj R-squared = 0.0123
Total | 2286.28937 9226 .247809383 Root MSE = .49473
------------------------------------------------------------------------------
college | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
distance | -.006371 .0005919 -10.76 0.000 -.0075312 -.0052108
_cons | .6091178 .0077287 78.81 0.000 .5939678 .6242678
------------------------------------------------------------------------------
Instrumental variables (2SLS) regression
Source | SS df MS Number of obs = 9227
-------------+------------------------------ F( 1, 9225) = 10.57
Model | 45.5110902 1 45.5110902 Prob > F = 0.0012
Residual | 1991.88243 9225 .215922215 R-squared = 0.0223
-------------+------------------------------ Adj R-squared = 0.0222
Total | 2037.39352 9226 .220831728 Root MSE = .46467
------------------------------------------------------------------------------
register | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
college | .2836913 .0872575 3.25 0.001 .1126472 .4547353
_cons | .5156526 .0479822 10.75 0.000 .4215969 .6097083
------------------------------------------------------------------------------
Instrumented: college
Instruments: distance
------------------------------------------------------------------------------
Table 10.4 on page 235.
reg3 (register=college) (college=distance), corr(u)
Three-stage least-squares regression
----------------------------------------------------------------------
Equation Obs Parms RMSE "R-sq" chi2 P
----------------------------------------------------------------------
register 9227 1 .4646239 0.0223 10.57 0.0011
college 9227 1 .4946807 0.0124 115.89 0.0000
----------------------------------------------------------------------
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
register |
college | .2836913 .0872481 3.25 0.001 .1126882 .4546944
_cons | .5156526 .047977 10.75 0.000 .4216194 .6096858
-------------+----------------------------------------------------------------
college |
distance | -.006371 .0005918 -10.77 0.000 -.0075309 -.005211
_cons | .6091178 .0077279 78.82 0.000 .5939714 .6242642
------------------------------------------------------------------------------
Endogenous variables: register college
Exogenous variables: distance
------------------------------------------------------------------------------
To estimate the error covariances, the following can be used immediately after the above
command.
* display matrix of error variances and covariances.
matrix list e(Sigma)
symmetric e(Sigma)[2,2]
register college
register .21587541
college -.02645365 .24470904
* calculate the correlation based on variances and covariances.
display -.02645365/(sqrt(.21587541)*sqrt(.24470904))
-.11509568
Table 10.5 on page 240.
ivreg register black hispanic otherrace (college=distance), first
First-stage regressions
-----------------------
Source | SS df MS Number of obs = 9227
-------------+------------------------------ F( 4, 9222) = 51.24
Model | 49.7051638 4 12.426291 Prob > F = 0.0000
Residual | 2236.5842 9222 .242527023 R-squared = 0.0217
-------------+------------------------------ Adj R-squared = 0.0213
Total | 2286.28937 9226 .247809383 Root MSE = .49247
------------------------------------------------------------------------------
college | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
black | -.057663 .015959 -3.61 0.000 -.0889461 -.0263799
hispanic | -.116213 .0132567 -8.77 0.000 -.1421991 -.0902269
otherrace | .0337076 .0240104 1.40 0.160 -.0133582 .0807734
distance | -.0069203 .0005947 -11.64 0.000 -.0080861 -.0057545
_cons | .643146 .0090535 71.04 0.000 .6253992 .6608929
------------------------------------------------------------------------------
Instrumental variables (2SLS) regression
Source | SS df MS Number of obs = 9227
-------------+------------------------------ F( 4, 9222) = 10.85
Model | 70.1890588 4 17.5472647 Prob > F = 0.0000
Residual | 1967.20446 9222 .213316467 R-squared = 0.0345
-------------+------------------------------ Adj R-squared = 0.0340
Total | 2037.39352 9226 .220831728 Root MSE = .46186
------------------------------------------------------------------------------
register | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
college | .2488616 .0806009 3.09 0.002 .0908659 .4068572
black | .061733 .0151736 4.07 0.000 .0319895 .0914766
hispanic | .0282926 .0148321 1.91 0.056 -.0007816 .0573669
otherrace | -.1066661 .0228193 -4.67 0.000 -.151397 -.0619352
_cons | .5266 .0463065 11.37 0.000 .435829 .6173709
------------------------------------------------------------------------------
Instrumented: college
Instruments: black hispanic otherrace distance
------------------------------------------------------------------------------
The analyses in Tables 10.6 and 10.7 include a variable that is restricted,
so these analyses cannot be reproduced using the available dataset.
Table 10.8 from page 249. Note that many of the coefficients in the first stage models
are equal to 0.
* Form interaction terms:
gen distxblack = distance*black
gen distxhispanic = distance*hispanic
gen distxotherrace = distance*otherrace
* Form interactions between instruments and exog race variables as instruments:
gen collegexblack = college*black
gen collegexhispanic = college*hispanic
gen collegexotherrace = college*otherrace
ivreg register black hispanic otherrace ///
(college collegexblack collegexhispanic collegexotherrace = ///
distance distxblack distxhispanic distxotherrace), first
First-stage regressions
-----------------------
Source | SS df MS Number of obs = 9227
-------------+------------------------------ F( 7, 9219) = 29.57
Model | 50.2079509 7 7.17256442 Prob > F = 0.0000
Residual | 2236.08142 9219 .242551407 R-squared = 0.0220
-------------+------------------------------ Adj R-squared = 0.0212
Total | 2286.28937 9226 .247809383 Root MSE = .4925
------------------------------------------------------------------------------
college | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
black | -.0650566 .0227625 -2.86 0.004 -.1096761 -.0204372
hispanic | -.12759 .0193679 -6.59 0.000 -.1655553 -.0896247
otherrace | .0589545 .0350741 1.68 0.093 -.0097986 .1277075
distance | -.0071118 .0007235 -9.83 0.000 -.00853 -.0056937
distxblack | .00089 .0020121 0.44 0.658 -.0030542 .0048342
distxhispa~c | .0012905 .0015768 0.82 0.413 -.0018003 .0043814
distxother~e | -.002994 .0029407 -1.02 0.309 -.0087584 .0027703
_cons | .6451829 .0100579 64.15 0.000 .6254673 .6648985
------------------------------------------------------------------------------
Source | SS df MS Number of obs = 9227
-------------+------------------------------ F( 7, 9219) = 1336.08
Model | 289.504372 7 41.3577675 Prob > F = 0.0000
Residual | 285.371102 9219 .03095467 R-squared = 0.5036
-------------+------------------------------ Adj R-squared = 0.5032
Total | 574.875474 9226 .06231037 Root MSE = .17594
------------------------------------------------------------------------------
collegexbl~k | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
black | .5801263 .0081317 71.34 0.000 .5641864 .5960661
hispanic | 1.71e-14 .006919 0.00 1.000 -.0135628 .0135628
otherrace | 5.19e-15 .0125299 0.00 1.000 -.0245614 .0245614
distance | 6.15e-16 .0002585 0.00 1.000 -.0005066 .0005066
distxblack | -.0062218 .0007188 -8.66 0.000 -.0076308 -.0048128
distxhispa~c | -1.53e-15 .0005633 -0.00 1.000 -.0011042 .0011042
distxother~e | -1.46e-15 .0010505 -0.00 1.000 -.0020593 .0020593
_cons | -3.37e-15 .0035931 -0.00 1.000 -.0070432 .0070432
------------------------------------------------------------------------------
Source | SS df MS Number of obs = 9227
-------------+------------------------------ F( 7, 9219) = 949.38
Model | 326.332652 7 46.6189502 Prob > F = 0.0000
Residual | 452.697586 9219 .049104847 R-squared = 0.4189
-------------+------------------------------ Adj R-squared = 0.4185
Total | 779.030237 9226 .084438569 Root MSE = .2216
------------------------------------------------------------------------------
collegexhi~c | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
black | -2.90e-14 .0102419 -0.00 1.000 -.0200763 .0200763
hispanic | .5175929 .0087145 59.39 0.000 .5005106 .5346753
otherrace | -6.01e-14 .0157815 -0.00 1.000 -.0309352 .0309352
distance | -2.64e-15 .0003255 -0.00 1.000 -.0006381 .0006381
distxblack | 8.00e-16 .0009053 0.00 1.000 -.0017747 .0017747
distxhispa~c | -.0058213 .0007095 -8.21 0.000 -.007212 -.0044306
distxother~e | 2.39e-15 .0013231 0.00 1.000 -.0025936 .0025936
_cons | 5.13e-14 .0045255 0.00 1.000 -.008871 .008871
------------------------------------------------------------------------------
Source | SS df MS Number of obs = 9227
-------------+------------------------------ F( 7, 9219) = 2119.51
Model | 168.032181 7 24.0045973 Prob > F = 0.0000
Residual | 104.410216 9219 .011325547 R-squared = 0.6168
-------------+------------------------------ Adj R-squared = 0.6165
Total | 272.442397 9226 .02952985 Root MSE = .10642
------------------------------------------------------------------------------
collegexot~e | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
black | 1.30e-14 .0049187 0.00 1.000 -.0096417 .0096417
hispanic | -2.71e-15 .0041851 -0.00 1.000 -.0082038 .0082038
otherrace | .7041374 .007579 92.91 0.000 .6892808 .718994
distance | 1.09e-16 .0001563 0.00 1.000 -.0003064 .0003064
distxblack | -1.31e-15 .0004348 -0.00 1.000 -.0008523 .0008523
distxhispa~c | 1.51e-17 .0003407 0.00 1.000 -.0006679 .0006679
distxother~e | -.0101059 .0006354 -15.90 0.000 -.0113515 -.0088603
_cons | 1.18e-15 .0021734 0.00 1.000 -.0042603 .0042603
------------------------------------------------------------------------------
Instrumental variables (2SLS) regression
Source | SS df MS Number of obs = 9227
-------------+------------------------------ F( 7, 9219) = 6.74
Model | 25.5540837 7 3.65058339 Prob > F = 0.0000
Residual | 2011.83944 9219 .218227512 R-squared = 0.0125
-------------+------------------------------ Adj R-squared = 0.0118
Total | 2037.39352 9226 .220831728 Root MSE = .46715
------------------------------------------------------------------------------
register | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
college | .3586927 .0964921 3.72 0.000 .1695469 .5478385
collegexbl~k | -.398586 .3020666 -1.32 0.187 -.9907034 .1935314
collegexhi~c | -.2929071 .247842 -1.18 0.237 -.7787322 .192918
collegexot~e | -.4633532 .2843963 -1.63 0.103 -1.020833 .0941266
black | .2780238 .1627012 1.71 0.088 -.0409066 .5969541
hispanic | .1765485 .120819 1.46 0.144 -.0602835 .4133804
otherrace | .1742172 .1755997 0.99 0.321 -.169997 .5184314
_cons | .4640456 .0552995 8.39 0.000 .3556464 .5724449
------------------------------------------------------------------------------
Instrumented: college collegexblack collegexhispanic collegexotherrace
Instruments: black hispanic otherrace distance distxblack distxhispanic
distxotherrace
------------------------------------------------------------------------------
Reproduce the analysis shown above (Table 10.8 page 249) using simultaneous equations.
Note that the results are identical. (Not shown in the text.)
reg3 (register = college collegexblack collegexhispanic collegexotherrace black hispanic otherrace) ///
(college = distance black hispanic otherrace distxblack distxhispanic distxotherrace) ///
(collegexblack = distance black hispanic otherrace distxblack distxhispanic distxotherrace) ///
(collegexhispanic = distance black hispanic otherrace distxblack distxhispanic distxotherrace) ///
(collegexotherrace = distance black hispanic otherrace distxblack distxhispanic distxotherrace), corr(u)
Three-stage least-squares regression
----------------------------------------------------------------------
Equation Obs Parms RMSE "R-sq" chi2 P
----------------------------------------------------------------------
register 9227 7 .4669457 0.0125 47.24 0.0000
college 9227 7 .4922815 0.0220 207.18 0.0000
collegexbl~k 9227 7 .1758631 0.5036 9360.64 0.0000
collegexhi~c 9227 7 .2215001 0.4189 6651.40 0.0000
collegexot~e 9227 7 .1063754 0.6168 14849.44 0.0000
----------------------------------------------------------------------
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
register |
college | .3586927 .0964502 3.72 0.000 .1696537 .5477316
collegexbl~k | -.398586 .3019356 -1.32 0.187 -.990369 .193197
collegexhi~c | -.2929071 .2477345 -1.18 0.237 -.7784578 .1926436
collegexot~e | -.4633532 .284273 -1.63 0.103 -1.020518 .0938117
black | .2780238 .1626307 1.71 0.087 -.0407265 .596774
hispanic | .1765485 .1207666 1.46 0.144 -.0601497 .4132467
otherrace | .1742172 .1755235 0.99 0.321 -.1698025 .518237
_cons | .4640456 .0552755 8.40 0.000 .3557076 .5723837
-------------+----------------------------------------------------------------
college |
distance | -.0071118 .0007232 -9.83 0.000 -.0085292 -.0056945
black | -.0650566 .0227526 -2.86 0.004 -.1096509 -.0204624
hispanic | -.12759 .0193595 -6.59 0.000 -.1655338 -.0896461
otherrace | .0589545 .0350589 1.68 0.093 -.0097597 .1276687
distxblack | .00089 .0020113 0.44 0.658 -.003052 .004832
distxhispa~c | .0012905 .0015761 0.82 0.413 -.0017986 .0043797
distxother~e | -.002994 .0029394 -1.02 0.308 -.0087551 .0027671
_cons | .6451829 .0100535 64.17 0.000 .6254784 .6648874
-------------+----------------------------------------------------------------
collegexbl~k |
distance | 1.96e-16 .0002583 0.00 1.000 -.0005063 .0005063
black | .5801263 .0081282 71.37 0.000 .5641954 .5960571
hispanic | 1.03e-14 .006916 0.00 1.000 -.0135551 .0135551
otherrace | 5.62e-15 .0125245 0.00 1.000 -.0245475 .0245475
distxblack | -.0062218 .0007185 -8.66 0.000 -.00763 -.0048136
distxhispa~c | -1.17e-15 .0005631 -0.00 1.000 -.0011036 .0011036
distxother~e | -6.23e-16 .0010501 -0.00 1.000 -.0020581 .0020581
_cons | -2.26e-15 .0035915 -0.00 1.000 -.0070393 .0070393
-------------+----------------------------------------------------------------
collegexhi~c |
distance | 2.05e-16 .0003254 0.00 1.000 -.0006377 .0006377
black | 3.62e-15 .0102374 0.00 1.000 -.020065 .020065
hispanic | .5175929 .0087107 59.42 0.000 .5005203 .5346656
otherrace | 5.26e-15 .0157746 0.00 1.000 -.0309177 .0309177
distxblack | -4.21e-16 .000905 -0.00 1.000 -.0017737 .0017737
distxhispa~c | -.0058213 .0007092 -8.21 0.000 -.0072112 -.0044314
distxother~e | -5.68e-16 .0013226 -0.00 1.000 -.0025922 .0025922
_cons | -1.90e-15 .0045235 -0.00 1.000 -.008866 .008866
-------------+----------------------------------------------------------------
collegexot~e |
distance | 7.94e-17 .0001563 0.00 1.000 -.0003063 .0003063
black | 1.33e-15 .0049165 0.00 1.000 -.0096362 .0096362
hispanic | 3.75e-15 .0041833 0.00 1.000 -.0081992 .0081992
otherrace | .7041374 .0075758 92.95 0.000 .6892892 .7189856
distxblack | -1.57e-16 .0004346 -0.00 1.000 -.0008518 .0008518
distxhispa~c | -4.19e-16 .0003406 -0.00 1.000 -.0006675 .0006675
distxother~e | -.0101059 .0006352 -15.91 0.000 -.0113507 -.008861
_cons | -7.91e-16 .0021724 -0.00 1.000 -.0042579 .0042579
------------------------------------------------------------------------------
Endogenous variables: register college collegexblack collegexhispanic
collegexotherrace
Exogenous variables: black hispanic otherrace distance distxblack
distxhispanic distxotherrace
------------------------------------------------------------------------------
matrix list e(Sigma)
symmetric e(Sigma)[5,5]
register college collegexbl~k collegexhi~c collegexot~e
register .2180383
college -.01156577 .24234111
collegexbl~k .00601057 .03092783 .03092783
collegexhi~c .00522595 .04906227 4.577e-34 .04906227
collegexot~e .00217381 .01131573 -1.553e-33 -1.915e-33 .01131573
Reproduce last two analyses (Table 10.8 page 249) using simultaneous equations, dropping
the irrelevant terms. Note that the results are identical. (Not shown in the text.)
reg3 (register = college collegexblack collegexhispanic collegexotherrace black hispanic otherrace) ///
(college = distance black hispanic otherrace distxblack distxhispanic distxotherrace) ///
(collegexblack = black distxblack) ///
(collegexhispanic = hispanic distxhispanic) ///
(collegexotherrace = otherrace distxotherrace), corr(u)
Three-stage least-squares regression
----------------------------------------------------------------------
Equation Obs Parms RMSE "R-sq" chi2 P
----------------------------------------------------------------------
register 9227 7 .4669457 0.0125 47.58 0.0000
college 9227 7 .4922815 0.0220 296.53 0.0000
collegexbl~k 9227 2 .1758631 0.5036 9360.64 0.0000
collegexhi~c 9227 2 .2215001 0.4189 6651.40 0.0000
collegexot~e 9227 2 .1063754 0.6168 14849.44 0.0000
----------------------------------------------------------------------
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
register |
college | .3586927 .0959752 3.74 0.000 .1705847 .5468006
collegexbl~k | -.398586 .3011776 -1.32 0.186 -.9888832 .1917112
collegexhi~c | -.2929071 .2467839 -1.19 0.235 -.7765947 .1907805
collegexot~e | -.4633532 .283115 -1.64 0.102 -1.018248 .091542
black | .2780238 .1622169 1.71 0.087 -.0399155 .5959631
hispanic | .1765485 .1202946 1.47 0.142 -.0592245 .4123215
otherrace | .1742172 .1748113 1.00 0.319 -.1684067 .5168411
_cons | .4640456 .0550057 8.44 0.000 .3562364 .5718549
-------------+----------------------------------------------------------------
college |
distance | -.0071118 .0005709 -12.46 0.000 -.0082308 -.0059929
black | -.0650566 .0194843 -3.34 0.001 -.1032451 -.0268682
hispanic | -.12759 .017194 -7.42 0.000 -.1612895 -.0938904
otherrace | .0589545 .0286356 2.06 0.040 .0028297 .1150792
distxblack | .00089 .0017235 0.52 0.606 -.002488 .0042681
distxhispa~c | .0012905 .0013947 0.93 0.355 -.0014431 .0040241
distxother~e | -.002994 .0024008 -1.25 0.212 -.0076995 .0017114
_cons | .6451829 .0086459 74.62 0.000 .6282372 .6621286
-------------+----------------------------------------------------------------
collegexbl~k |
black | .5801263 .0075499 76.84 0.000 .5653288 .5949237
distxblack | -.0062218 .0006705 -9.28 0.000 -.0075359 -.0049077
_cons | -6.09e-18 .0019577 -0.00 1.000 -.0038369 .0038369
-------------+----------------------------------------------------------------
collegexhi~c |
hispanic | .5175929 .0078773 65.71 0.000 .5021538 .5330321
distxhispa~c | -.0058213 .0006301 -9.24 0.000 -.0070563 -.0045863
_cons | -3.52e-17 .0025763 -0.00 1.000 -.0050494 .0050494
-------------+----------------------------------------------------------------
collegexot~e |
otherrace | .7041374 .0073459 95.85 0.000 .6897396 .7185351
distxother~e | -.0101059 .0006156 -16.42 0.000 -.0113125 -.0088992
_cons | -2.26e-17 .0011358 -0.00 1.000 -.0022261 .0022261
------------------------------------------------------------------------------
Endogenous variables: register college collegexblack collegexhispanic
collegexotherrace
Exogenous variables: black hispanic otherrace distance distxblack
distxhispanic distxotherrace
------------------------------------------------------------------------------
matrix list e(Sigma)
symmetric e(Sigma)[5,5]
register college collegexbl~k collegexhi~c collegexot~e
register .2180383
college -.01156577 .24234111
collegexbl~k .00601057 .03092783 .03092783
collegexhi~c .00522595 .04906227 0 .04906227
collegexot~e .00217381 .01131573 0 0 .01131573
A model similar to the last three (all of which replicate Table 10.8 page 249) estimated using
simultaneous equations, dropping
the irrelevant terms and non-significant variables from the model. Note that the results are very similar, but
not identical. (Not shown in the text.)
reg3 (register = college collegexblack collegexhispanic collegexotherrace black hisp otherrace) ///
(college = distance black hispanic otherrace) ///
(collegexblack = black distxblack) ///
(collegexhispanic = hispanic distxhispanic) ///
(collegexotherrace = otherrace distxotherrace), corr(u)
Three-stage least-squares regression
----------------------------------------------------------------------
Equation Obs Parms RMSE "R-sq" chi2 P
----------------------------------------------------------------------
register 9227 7 .4673456 0.0109 47.78 0.0000
college 9227 4 .4923371 0.0217 293.54 0.0000
collegexbl~k 9227 2 .1758635 0.5036 9380.65 0.0000
collegexhi~c 9227 2 .2215025 0.4189 6684.76 0.0000
collegexot~e 9227 2 .1063761 0.6168 14856.35 0.0000
----------------------------------------------------------------------
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
register |
college | .3618827 .09572 3.78 0.000 .174275 .5494905
collegexbl~k | -.4143828 .2996661 -1.38 0.167 -1.001718 .172952
collegexhi~c | -.3156465 .2454996 -1.29 0.199 -.7968169 .1655239
collegexot~e | -.4230397 .2809778 -1.51 0.132 -.9737461 .1276667
black | .2865527 .1614041 1.78 0.076 -.0297935 .6028988
hispanic | .1875168 .119659 1.57 0.117 -.0470105 .4220441
otherrace | .1491671 .1735122 0.86 0.390 -.1909106 .4892448
_cons | .4622287 .0548613 8.43 0.000 .3547026 .5697549
-------------+----------------------------------------------------------------
college |
distance | -.0069746 .0004853 -14.37 0.000 -.0079257 -.0060235
black | -.057824 .0137609 -4.20 0.000 -.0847949 -.0308531
hispanic | -.116329 .0119552 -9.73 0.000 -.1397607 -.0928972
otherrace | .0335953 .019631 1.71 0.087 -.0048807 .0720712
_cons | .6437234 .0080333 80.13 0.000 .6279783 .6594684
-------------+----------------------------------------------------------------
collegexbl~k |
black | .5811072 .0072738 79.89 0.000 .5668508 .5953636
distxblack | -.0063497 .0006164 -10.30 0.000 -.0075578 -.0051416
_cons | -3.56e-17 .0019577 -0.00 1.000 -.0038369 .0038369
-------------+----------------------------------------------------------------
collegexhi~c |
hispanic | .5199955 .0074862 69.46 0.000 .5053227 .5346683
distxhispa~c | -.006104 .0005602 -10.90 0.000 -.007202 -.0050059
_cons | -3.05e-17 .0025763 -0.00 1.000 -.0050494 .0050494
-------------+----------------------------------------------------------------
collegexot~e |
otherrace | .7022688 .0072187 97.28 0.000 .6881204 .7164171
distxother~e | -.0098877 .0005948 -16.62 0.000 -.0110534 -.008722
_cons | 3.47e-18 .0011358 0.00 1.000 -.0022261 .0022261
------------------------------------------------------------------------------
Endogenous variables: register college collegexblack collegexhispanic
collegexotherrace
Exogenous variables: black hispanic otherrace distance distxblack
distxhispanic distxotherrace
------------------------------------------------------------------------------
matrix list e(Sigma)
symmetric e(Sigma)[5,5]
register college collegexbl~k collegexhi~c collegexot~e
register .2180383
college -.01156577 .2423956
collegexbl~k .00601057 .03092783 .03092783
collegexhi~c .00522595 .04906227 0 .04906227
collegexot~e .00217381 .01131573 0 0 .01131573
The analyses in shown in Table 10.9 includes a variable that is restricted, so these
analyses cannot be reproduced using the available dataset.