|
|
|
||||
|
|
|||||
Note: The generate command is used to add an id number to the observations.
use http://www.ats.ucla.edu/stat/stata/examples/chp/p124 generate id = _n
Table 5.1, page 124.
list
s x e m
1. 13876 1 1 1
2. 11608 1 3 0
3. 18701 1 3 1
4. 11283 1 2 0
5. 11767 1 3 0
6. 20872 2 2 1
7. 11772 2 2 0
8. 10535 2 1 0
9. 12195 2 3 0
10. 12313 3 2 0
..
[remainder of output omitted]
Create dummy coding for variable e.
Note 1: tabulate with the generate option creates dummy coded variables.
Note 2: The tab1 command produces one-way frequency tables for a series of variables.
tabulate e, generate(e)
E | Freq. Percent Cum.
------------+-----------------------------------
1 | 14 30.43 30.43
2 | 19 41.30 71.74
3 | 13 28.26 100.00
------------+-----------------------------------
Total | 46 100.00
tab1 e1 e2 e3
-> tabulation of e1
e== |
1.0000 | Freq. Percent Cum.
------------+-----------------------------------
0 | 32 69.57 69.57
1 | 14 30.43 100.00
------------+-----------------------------------
Total | 46 100.00
-> tabulation of e2
e== |
2.0000 | Freq. Percent Cum.
------------+-----------------------------------
0 | 27 58.70 58.70
1 | 19 41.30 100.00
------------+-----------------------------------
Total | 46 100.00
-> tabulation of e3
e== |
3.0000 | Freq. Percent Cum.
------------+-----------------------------------
0 | 33 71.74 71.74
1 | 13 28.26 100.00
------------+-----------------------------------
Total | 46 100.00
Table 5.3, page 126.
regress s x e1 e2 m
Source | SS df MS Number of obs = 46
---------+------------------------------ F( 4, 41) = 226.84
Model | 957816858 4 239454214 Prob > F = 0.0000
Residual | 43280719.5 41 1055627.30 R-squared = 0.9568
---------+------------------------------ Adj R-squared = 0.9525
Total | 1.0011e+09 45 22246612.8 Root MSE = 1027.4
------------------------------------------------------------------------------
s | Coef. Std. Err. t P>|t| [95% Conf. Interval]
---------+--------------------------------------------------------------------
x | 546.184 30.51919 17.896 0.000 484.5493 607.8188
e1 | -2996.21 411.7527 -7.277 0.000 -3827.762 -2164.659
e2 | 147.8249 387.6593 0.381 0.705 -635.0689 930.7188
m | 6883.531 313.919 21.928 0.000 6249.559 7517.503
_cons | 11031.81 383.2171 28.787 0.000 10257.89 11805.73
------------------------------------------------------------------------------
Figure 5.1, page 127. The rvpplot2 can be downloaded within Stata by typing findit rvpplot2 (see How can I use the findit command to search for programs and get additional help? for more information about using findit).
rvpplot2 x, rstudent xlabel(4(4)20) ylabel(-2(1)1)
Figure 5.2, page 128.
Note 1: The egen command was used with the group function to produce the six categories of education and management.
Note 2: The xlabel(1(1)6) option produces labels on the x-axis from 1 to 6 by ones.
egen c = group(e m) predict sr, rstandard graph twoway scatter sr c, ylabel(-2(1)1) xlabel(1(1)6)
Table 5.4, page 129.
generate e1m = e1*m
generate e2m = e2*m
regress s x e1 e2 m e1m e2m
Source | SS df MS Number of obs = 46
---------+------------------------------ F( 6, 39) = 5516.60
Model | 999919409 6 166653235 Prob > F = 0.0000
Residual | 1178167.86 39 30209.4324 R-squared = 0.9988
---------+------------------------------ Adj R-squared = 0.9986
Total | 1.0011e+09 45 22246612.8 Root MSE = 173.81
------------------------------------------------------------------------------
s | Coef. Std. Err. t P>|t| [95% Conf. Interval]
---------+--------------------------------------------------------------------
x | 496.987 5.566415 89.283 0.000 485.7279 508.2461
e1 | -1730.748 105.3339 -16.431 0.000 -1943.806 -1517.69
e2 | -349.0777 97.5679 -3.578 0.001 -546.4274 -151.728
m | 7047.412 102.5892 68.695 0.000 6839.906 7254.918
e1m | -3066.035 149.3304 -20.532 0.000 -3368.084 -2763.986
e2m | 1836.488 131.1674 14.001 0.000 1571.177 2101.799
_cons | 11203.43 79.06545 141.698 0.000 11043.51 11363.36
------------------------------------------------------------------------------
Figure 5.3, page 129.
rvpplot2 x, rstandard xlabel(4(4)20) ylabel(-6(1.5)1.5)
Table 5.5, page 129.
drop if id==33
(1 observation deleted)
regress s x e1 e2 m e1m e2m
Source | SS df MS Number of obs = 45
---------+------------------------------ F( 6, 38) =35427.96
Model | 957607113 6 159601186 Prob > F = 0.0000
Residual | 171188.12 38 4504.95052 R-squared = 0.9998
---------+------------------------------ Adj R-squared = 0.9998
Total | 957778301 44 21767688.7 Root MSE = 67.119
------------------------------------------------------------------------------
s | Coef. Std. Err. t P>|t| [95% Conf. Interval]
---------+--------------------------------------------------------------------
x | 498.4178 2.151688 231.640 0.000 494.0619 502.7736
e1 | -1741.336 40.6825 -42.803 0.000 -1823.693 -1658.979
e2 | -357.0423 37.68114 -9.475 0.000 -433.3237 -280.7608
m | 7040.58 39.61907 177.707 0.000 6960.376 7120.785
e1m | -3051.763 57.6742 -52.914 0.000 -3168.519 -2935.008
e2m | 1997.531 51.78498 38.574 0.000 1892.697 2102.364
_cons | 11199.71 30.53338 366.802 0.000 11137.9 11261.53
------------------------------------------------------------------------------
Figure 5.4, page 130.
rvpplot2 x, rstandard xlabel(4(4)20) ylabel(-2(1)2)
Figure 5.5, page 130.
drop sr predict sr, rstandard graph twoway scatter sr c, ylabel(-2(1)1) xlabel(1(1)6)
Table 5.6, page 131.
Note: There are small differences between table 5.6 in the book and the one generated by Stata due to rounding.
adjust x=0, by(m e) se ci format(%5.0f)
-------------------------------------------------------------------------------
Dependent variable: s Command: regress
Variables left as is: e1, e2, e1m, e2m
Covariate set to value: x = 0
-------------------------------------------------------------------------------
----------+--------------------------------------------
| E
M | 1 2 3
----------+--------------------------------------------
0 | 9458 10843 11200
| (31) (26) (31)
| [9396,9521] [10790,10896] [11138,11262]
|
1 | 13447 19881 18240
| (32) (33) (29)
| [13383,13511] [19814,19947] [18183,18298]
----------+--------------------------------------------
Use the data file p134 and create interaction variable.
use http://www.ats.ucla.edu/stat/stata/examples/chp/p134, clear generate rbyt = race*test
Table 5.8, page 135.
regress jperf test
Source | SS df MS Number of obs = 20
---------+------------------------------ F( 1, 18) = 19.25
Model | 48.7229625 1 48.7229625 Prob > F = 0.0004
Residual | 45.5682959 18 2.531572 R-squared = 0.5167
---------+------------------------------ Adj R-squared = 0.4899
Total | 94.2912585 19 4.96269781 Root MSE = 1.5911
------------------------------------------------------------------------------
jperf | Coef. Std. Err. t P>|t| [95% Conf. Interval]
---------+--------------------------------------------------------------------
test | 2.360535 .5380699 4.387 0.000 1.230092 3.490978
_cons | 1.034973 .8680312 1.192 0.249 -.7886928 2.858639
------------------------------------------------------------------------------
Figure 5.7, page 135.
rvpplot2 test, rstandard ylabel(-2(1)1) xlabel(.75 1.5 2.25)
Figure 5.9, page 137.
predict sr, rstandard graph twoway scatter sr race, ylabel(-2(1)1) xlabel(0 1)
Table 5.9, page 135.
regress jperf test race rbyt
Source | SS df MS Number of obs = 20
---------+------------------------------ F( 3, 16) = 10.55
Model | 62.6357847 3 20.8785949 Prob > F = 0.0005
Residual | 31.6554738 16 1.97846711 R-squared = 0.6643
---------+------------------------------ Adj R-squared = 0.6013
Total | 94.2912585 19 4.96269781 Root MSE = 1.4066
------------------------------------------------------------------------------
jperf | Coef. Std. Err. t P>|t| [95% Conf. Interval]
---------+--------------------------------------------------------------------
test | 1.313402 .6703711 1.959 0.068 -.1077208 2.734525
race | -1.913167 1.540325 -1.242 0.232 -5.178509 1.352176
rbyt | 1.997546 .954443 2.093 0.053 -.0257831 4.020875
_cons | 2.010282 1.050112 1.914 0.074 -.2158562 4.236421
------------------------------------------------------------------------------
Figure 5.8, page 135.
rvpplot2 test, rstandard ylabel(-2(1)1) xlabel(.75 1.5 2.25)
Part of Table 5.10, page 136.
regress jperf test if race==1
Source | SS df MS Number of obs = 10
---------+------------------------------ F( 1, 8) = 28.14
Model | 46.9895716 1 46.9895716 Prob > F = 0.0007
Residual | 13.3568411 8 1.66960513 R-squared = 0.7787
---------+------------------------------ Adj R-squared = 0.7510
Total | 60.3464126 9 6.70515696 Root MSE = 1.2921
------------------------------------------------------------------------------
jperf | Coef. Std. Err. t P>|t| [95% Conf. Interval]
---------+--------------------------------------------------------------------
test | 3.310948 .6241062 5.305 0.001 1.871757 4.75014
_cons | .0971152 1.035193 0.094 0.928 -2.290043 2.484274
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Figure 5.10, page 137.
rvpplot2 test, rstandard ylabel(-1.5(.75).75) xlabel(.75 1.5 2.25)
Remainder of Table 5.10, page 136.
regress jperf test if race==0
Source | SS df MS Number of obs = 10
---------+------------------------------ F( 1, 8) = 3.32
Model | 7.5944073 1 7.5944073 Prob > F = 0.1059
Residual | 18.2986327 8 2.28732909 R-squared = 0.2933
---------+------------------------------ Adj R-squared = 0.2050
Total | 25.89304 9 2.87700445 Root MSE = 1.5124
------------------------------------------------------------------------------
jperf | Coef. Std. Err. t P>|t| [95% Conf. Interval]
---------+--------------------------------------------------------------------
test | 1.313402 .7208006 1.822 0.106 -.3487669 2.975572
_cons | 2.010282 1.129108 1.780 0.113 -.5934463 4.614011
------------------------------------------------------------------------------
Figure 5.11, page 137.
rvpplot2 test, rstandard ylabel(-1.5(.75).75) xlabel(.8(.4)2.4)
UCLA Researchers are invited to our Statistical Consulting Services
We recommend others to our list of Other Resources for Statistical Computing Help
These pages are Copyrighted (c) by UCLA Academic Technology Services
