------------------------------------------------------------------------------------------------------------------------------------------------------
name: STATA_Chapter9
log: C:\Dropbox\PilesOfVariance\Chapter9\STATA\STATA_Chapter9_Output.smcl
log type: smcl
opened on: 20 Jan 2015, 16:07:35
.
. display as result "Chapter 9: Descriptive Statistics for Time-Varying Variables"
Chapter 9: Descriptive Statistics for Time-Varying Variables
. preserve
. collapse attitude12 pmmonitor copymonitor12 copymonitor18, by(personid)
. summarize attitude12 pmmonitor copymonitor12 copymonitor18
Variable | Obs Mean Std. Dev. Min Max
-------------+--------------------------------------------------------
attitude12 | 200 3.9505 .6024956 2.437367 5
pmmonitor | 200 3.075439 .5526113 1.153501 4.47294
copymonit~12 | 200 3.078307 .8040337 1 5
copymonit~18 | 200 3.067673 .5561235 1.339776 4.420773
. corr pmmonitor copymonitor12 copymonitor18
(obs=200)
| pmmoni~r copym~12 copym~18
-------------+---------------------------
pmmonitor | 1.0000
copymonit~12 | 0.9137 1.0000
copymonit~18 | 0.7401 0.5246 1.0000
. restore
.
. display as result "Chapter 9: Descriptive Statistics for Time-Varying Variables"
Chapter 9: Descriptive Statistics for Time-Varying Variables
. summarize age risky monitor wpmon change18mon
Variable | Obs Mean Std. Dev. Min Max
-------------+--------------------------------------------------------
age | 1400 15.00103 2.004174 11.53088 18.33688
risky | 1400 19.38489 5.301359 10 36.28243
monitor | 1400 3.075439 .6445657 1 5
wpmon | 1400 1.91e-09 .3337595 -1.136053 1.226129
change18mon | 1400 .0077664 .5200831 -2.355284 1.713193
. corr pmmon3 age18mon3 wpmon change18mon
(obs=1400)
| pmmon3 age18m~3 wpmon change~n
-------------+------------------------------------
pmmon3 | 1.0000
age18mon3 | 0.7401 1.0000
wpmon | 0.0000 0.0000 1.0000
change18mon | 0.2706 -0.2823 0.6417 1.0000
.
. display as result "Ch 9: Empty Means, Random Intercept Model for Monitoring"
Ch 9: Empty Means, Random Intercept Model for Monitoring
. mixed monitor , ///
> || personid: , variance mle covariance(unstructured),
Note: single-variable random-effects specification in personid equation; covariance structure set to identity
Performing EM optimization:
Performing gradient-based optimization:
Iteration 0: log likelihood = -837.24375
Iteration 1: log likelihood = -837.24375
Computing standard errors:
Mixed-effects ML regression Number of obs = 1400
Group variable: personid Number of groups = 200
Obs per group: min = 7
avg = 7.0
max = 7
Wald chi2(0) = .
Log likelihood = -837.24375 Prob > chi2 = .
------------------------------------------------------------------------------
monitor | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_cons | 3.075439 .0389777 78.90 0.000 2.999044 3.151834
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
personid: Identity |
var(_cons) | .2852998 .0303947 .2315357 .3515483
-----------------------------+------------------------------------------------
var(Residual) | .1298685 .0053019 .1198819 .140687
------------------------------------------------------------------------------
LR test vs. linear regression: chibar2(01) = 1067.84 Prob >= chibar2 = 0.0000
. estat ic, n(200),
Akaike's information criterion and Bayesian information criterion
-----------------------------------------------------------------------------
Model | Obs ll(null) ll(model) df AIC BIC
-------------+---------------------------------------------------------------
. | 200 . -837.2438 3 1680.488 1690.382
-----------------------------------------------------------------------------
Note: N=200 used in calculating BIC
. estat icc,
Intraclass correlation
------------------------------------------------------------------------------
Level | ICC Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
personid | .6871908 .0247281 .6368114 .7335062
------------------------------------------------------------------------------
. estat wcorrelation, covariance,
Covariances for personid = 1:
obs | 1 2 3 4 5 6 7
-------------+--------------------------------------------------------
1 | 0.415
2 | 0.285 0.415
3 | 0.285 0.285 0.415
4 | 0.285 0.285 0.285 0.415
5 | 0.285 0.285 0.285 0.285 0.415
6 | 0.285 0.285 0.285 0.285 0.285 0.415
7 | 0.285 0.285 0.285 0.285 0.285 0.285 0.415
. estat wcorrelation,
Standard deviations and correlations for personid = 1:
Standard deviations:
obs | 1 2 3 4 5 6 7
-------------+--------------------------------------------------------
sd | 0.644 0.644 0.644 0.644 0.644 0.644 0.644
Correlations:
obs | 1 2 3 4 5 6 7
-------------+--------------------------------------------------------
1 | 1.000
2 | 0.687 1.000
3 | 0.687 0.687 1.000
4 | 0.687 0.687 0.687 1.000
5 | 0.687 0.687 0.687 0.687 1.000
6 | 0.687 0.687 0.687 0.687 0.687 1.000
7 | 0.687 0.687 0.687 0.687 0.687 0.687 1.000
.
. display as result "Ch 9: Fixed Linear Age, Random Intercept Model for Monitoring"
Ch 9: Fixed Linear Age, Random Intercept Model for Monitoring
. mixed monitor c.agec18, ///
> || personid: , variance mle covariance(unstructured),
Note: single-variable random-effects specification in personid equation; covariance structure set to identity
Performing EM optimization:
Performing gradient-based optimization:
Iteration 0: log likelihood = -837.0184
Iteration 1: log likelihood = -837.0184
Computing standard errors:
Mixed-effects ML regression Number of obs = 1400
Group variable: personid Number of groups = 200
Obs per group: min = 7
avg = 7.0
max = 7
Wald chi2(1) = 0.45
Log likelihood = -837.0184 Prob > chi2 = 0.5020
------------------------------------------------------------------------------
monitor | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
agec18 | -.0032289 .0048092 -0.67 0.502 -.0126547 .0061969
_cons | 3.065756 .0415588 73.77 0.000 2.984302 3.14721
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
personid: Identity |
var(_cons) | .2852791 .030392 .2315197 .3515216
-----------------------------+------------------------------------------------
var(Residual) | .1298216 .0052999 .1198387 .1406362
------------------------------------------------------------------------------
LR test vs. linear regression: chibar2(01) = 1068.05 Prob >= chibar2 = 0.0000
. estat ic, n(200),
Akaike's information criterion and Bayesian information criterion
-----------------------------------------------------------------------------
Model | Obs ll(null) ll(model) df AIC BIC
-------------+---------------------------------------------------------------
. | 200 . -837.0184 4 1682.037 1695.23
-----------------------------------------------------------------------------
Note: N=200 used in calculating BIC
. estimates store FitFixLin,
.
. display as result "Ch 9: Random Linear Age Model for Monitoring"
Ch 9: Random Linear Age Model for Monitoring
. mixed monitor c.agec18, ///
> || personid: agec18, variance mle covariance(unstructured),
Performing EM optimization:
Performing gradient-based optimization:
Iteration 0: log likelihood = -676.87612
Iteration 1: log likelihood = -676.87612
Computing standard errors:
Mixed-effects ML regression Number of obs = 1400
Group variable: personid Number of groups = 200
Obs per group: min = 7
avg = 7.0
max = 7
Wald chi2(1) = 0.16
Log likelihood = -676.87612 Prob > chi2 = 0.6861
------------------------------------------------------------------------------
monitor | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
agec18 | -.0033057 .0081793 -0.40 0.686 -.0193369 .0127255
_cons | 3.065015 .034128 89.81 0.000 2.998126 3.131905
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
personid: Unstructured |
var(agec18) | .0104944 .0013447 .0081638 .0134905
var(_cons) | .1954436 .0233316 .1546704 .2469652
cov(agec18,_cons) | -.000423 .0040081 -.0082787 .0074326
-----------------------------+------------------------------------------------
var(Residual) | .0807506 .0036117 .0739732 .0881489
------------------------------------------------------------------------------
LR test vs. linear regression: chi2(3) = 1388.33 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
. estat ic, n(200),
Akaike's information criterion and Bayesian information criterion
-----------------------------------------------------------------------------
Model | Obs ll(null) ll(model) df AIC BIC
-------------+---------------------------------------------------------------
. | 200 . -676.8761 6 1365.752 1385.542
-----------------------------------------------------------------------------
Note: N=200 used in calculating BIC
. estimates store FitRandLin,
. lrtest FitRandLin FitFixLin,
Likelihood-ratio test LR chi2(2) = 320.28
(Assumption: FitFixLin nested in FitRandLin) Prob > chi2 = 0.0000
Note: The reported degrees of freedom assumes the null hypothesis is not on the boundary of the parameter space. If this is not true, then the
reported test is conservative.
.
. display as result "Ch 9: Fixed Quadratic, Random Linear Age Model for Monitoring"
Ch 9: Fixed Quadratic, Random Linear Age Model for Monitoring
. mixed monitor c.agec18 c.agec18#c.agec18, ///
> || personid: agec18, variance mle covariance(unstructured),
Performing EM optimization:
Performing gradient-based optimization:
Iteration 0: log likelihood = -676.86623
Iteration 1: log likelihood = -676.86623
Computing standard errors:
Mixed-effects ML regression Number of obs = 1400
Group variable: personid Number of groups = 200
Obs per group: min = 7
avg = 7.0
max = 7
Wald chi2(2) = 0.18
Log likelihood = -676.86623 Prob > chi2 = 0.9125
-----------------------------------------------------------------------------------
monitor | Coef. Std. Err. z P>|z| [95% Conf. Interval]
------------------+----------------------------------------------------------------
agec18 | -.0051289 .0153292 -0.33 0.738 -.0351736 .0249158
|
c.agec18#c.agec18 | -.0003039 .0021611 -0.14 0.888 -.0045396 .0039318
|
_cons | 3.063499 .0357907 85.59 0.000 2.993351 3.133648
-----------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
personid: Unstructured |
var(agec18) | .0104947 .0013447 .008164 .0134908
var(_cons) | .1954453 .0233316 .154672 .246967
cov(agec18,_cons) | -.0004243 .0040081 -.00828 .0074314
-----------------------------+------------------------------------------------
var(Residual) | .0807485 .0036116 .0739713 .0881466
------------------------------------------------------------------------------
LR test vs. linear regression: chi2(3) = 1388.29 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
. estat ic, n(200),
Akaike's information criterion and Bayesian information criterion
-----------------------------------------------------------------------------
Model | Obs ll(null) ll(model) df AIC BIC
-------------+---------------------------------------------------------------
. | 200 . -676.8662 7 1367.732 1390.821
-----------------------------------------------------------------------------
Note: N=200 used in calculating BIC
. estimates store FitFixQuad,
.
. display as result "Ch 9: Random Quadratic Age Model for Monitoring"
Ch 9: Random Quadratic Age Model for Monitoring
. mixed monitor c.agec18 c.agec18#c.agec18, ///
> || personid: agec18 agec18sq, variance mle covariance(unstructured),
Performing EM optimization:
Performing gradient-based optimization:
Iteration 0: log likelihood = -675.72747
Iteration 1: log likelihood = -674.36638
Iteration 2: log likelihood = -674.17384
Iteration 3: log likelihood = -674.12505
Iteration 4: log likelihood = -674.12064
Iteration 5: log likelihood = -674.12021
Iteration 6: log likelihood = -674.1202
Computing standard errors:
Mixed-effects ML regression Number of obs = 1400
Group variable: personid Number of groups = 200
Obs per group: min = 7
avg = 7.0
max = 7
Wald chi2(2) = 0.19
Log likelihood = -674.1202 Prob > chi2 = 0.9076
-----------------------------------------------------------------------------------
monitor | Coef. Std. Err. z P>|z| [95% Conf. Interval]
------------------+----------------------------------------------------------------
agec18 | -.0060175 .0157907 -0.38 0.703 -.0369667 .0249317
|
c.agec18#c.agec18 | -.0004637 .0021911 -0.21 0.832 -.0047581 .0038308
|
_cons | 3.062668 .0374521 81.78 0.000 2.989263 3.136072
-----------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
personid: Unstructured |
var(agec18) | .013574 . . .
var(agec18sq) | .000032 . . .
var(_cons) | .2200984 . . .
cov(agec18,agec18sq) | .000349 . . .
cov(agec18,_cons) | .0157988 . . .
cov(agec18sq,_cons) | .0025617 . . .
-----------------------------+------------------------------------------------
var(Residual) | .0801881 . . .
------------------------------------------------------------------------------
LR test vs. linear regression: chi2(6) = 1393.78 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
. estat ic, n(200),
Akaike's information criterion and Bayesian information criterion
-----------------------------------------------------------------------------
Model | Obs ll(null) ll(model) df AIC BIC
-------------+---------------------------------------------------------------
. | 200 . -674.1202 3 1354.24 1364.135
-----------------------------------------------------------------------------
Note: N=200 used in calculating BIC
. estimates store FitRandQuad,
. lrtest FitRandQuad FitFixQuad, force
Likelihood-ratio test LR chi2(4) = -5.49
(Assumption: FitRandQuad nested in FitFixQuad) Prob > chi2 = 1.0000
.
. display as result "Ch 9: Fixed Quadratic, Random Linear Age Model for Risky Behavior"
Ch 9: Fixed Quadratic, Random Linear Age Model for Risky Behavior
. display as result "Conditional Baseline with Attitudes Predicting Linear Age Slope"
Conditional Baseline with Attitudes Predicting Linear Age Slope
. mixed risky c.agec18 c.agec18#c.agec18 ///
> c.att4 c.agec18#c.att4, ///
> || personid: agec18, variance mle covariance(unstructured),
Performing EM optimization:
Performing gradient-based optimization:
Iteration 0: log likelihood = -3801.2664
Iteration 1: log likelihood = -3801.2664
Computing standard errors:
Mixed-effects ML regression Number of obs = 1400
Group variable: personid Number of groups = 200
Obs per group: min = 7
avg = 7.0
max = 7
Wald chi2(4) = 397.26
Log likelihood = -3801.2664 Prob > chi2 = 0.0000
-----------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
------------------+----------------------------------------------------------------
agec18 | 1.964886 .1460116 13.46 0.000 1.678709 2.251064
|
c.agec18#c.agec18 | .1450475 .021963 6.60 0.000 .1020007 .1880943
|
att4 | -3.155502 .5512883 -5.72 0.000 -4.236007 -2.074997
|
c.agec18#c.att4 | -.5154258 .1042977 -4.94 0.000 -.7198455 -.3110061
|
_cons | 23.31201 .350051 66.60 0.000 22.62592 23.9981
-----------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
personid: Unstructured |
var(agec18) | .4880961 .0798137 .3542553 .672503
var(_cons) | 18.08098 2.204655 14.23748 22.96206
cov(agec18,_cons) | 1.884828 .3564732 1.186154 2.583503
-----------------------------+------------------------------------------------
var(Residual) | 8.352883 .373583 7.651849 9.118143
------------------------------------------------------------------------------
LR test vs. linear regression: chi2(3) = 663.34 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
. estat ic, n(200),
Akaike's information criterion and Bayesian information criterion
-----------------------------------------------------------------------------
Model | Obs ll(null) ll(model) df AIC BIC
-------------+---------------------------------------------------------------
. | 200 . -3801.266 9 7620.533 7650.218
-----------------------------------------------------------------------------
Note: N=200 used in calculating BIC
. predict PredAttOnly, xb,
.
. display as result "Eq 9.1: Predicting Quadratic Change in Risky Behavior"
Eq 9.1: Predicting Quadratic Change in Risky Behavior
. display as result "From Person Mean Monitoring as Between-Person Monitoring"
From Person Mean Monitoring as Between-Person Monitoring
. mixed risky c.agec18 c.agec18#c.agec18 c.att4 c.agec18#c.att4 ///
> c.pmmon3 c.agec18#c.pmmon3 c.agec18#c.agec18#c.pmmon3, ///
> || personid: agec18, variance mle covariance(unstructured),
Performing EM optimization:
Performing gradient-based optimization:
Iteration 0: log likelihood = -3779.776
Iteration 1: log likelihood = -3779.776
Computing standard errors:
Mixed-effects ML regression Number of obs = 1400
Group variable: personid Number of groups = 200
Obs per group: min = 7
avg = 7.0
max = 7
Wald chi2(7) = 452.39
Log likelihood = -3779.776 Prob > chi2 = 0.0000
--------------------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
---------------------------+----------------------------------------------------------------
agec18 | 1.925601 .1470975 13.09 0.000 1.637295 2.213907
|
c.agec18#c.agec18 | .1362197 .0221706 6.14 0.000 .0927661 .1796733
|
att4 | -3.31951 .5141918 -6.46 0.000 -4.327307 -2.311712
|
c.agec18#c.att4 | -.5237898 .1037545 -5.05 0.000 -.727145 -.3204347
|
pmmon3 | -2.590697 .5924507 -4.37 0.000 -3.751879 -1.429515
|
c.agec18#c.pmmon3 | .4454477 .2618253 1.70 0.089 -.0677203 .9586158
|
c.agec18#c.agec18#c.pmmon3 | .1037089 .0395334 2.62 0.009 .0262248 .1811929
|
_cons | 23.49447 .3312389 70.93 0.000 22.84525 24.14368
--------------------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
personid: Unstructured |
var(agec18) | .4793567 .0787589 .3473805 .6614732
var(_cons) | 15.18712 1.913957 11.86322 19.44232
cov(agec18,_cons) | 1.722833 .3317575 1.0726 2.373066
-----------------------------+------------------------------------------------
var(Residual) | 8.299335 .3711739 7.602821 9.05966
------------------------------------------------------------------------------
LR test vs. linear regression: chi2(3) = 563.42 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
. estat ic, n(200),
Akaike's information criterion and Bayesian information criterion
-----------------------------------------------------------------------------
Model | Obs ll(null) ll(model) df AIC BIC
-------------+---------------------------------------------------------------
. | 200 . -3779.776 12 7583.552 7623.132
-----------------------------------------------------------------------------
Note: N=200 used in calculating BIC
. * Effect of PM Monitoring at Age 12
. lincom pmmon3*1 + c.agec18#c.pmmon3*-6 + c.agec18#c.agec18#c.pmmon3*36
( 1) [risky]pmmon3 - 6*[risky]c.agec18#c.pmmon3 + 36*[risky]c.agec18#c.agec18#c.pmmon3 = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | -1.529865 .5449359 -2.81 0.005 -2.597919 -.4618097
------------------------------------------------------------------------------
. * Effect of PM Monitoring at Age 14
. lincom pmmon3*1 + c.agec18#c.pmmon3*-4 + c.agec18#c.agec18#c.pmmon3*16
( 1) [risky]pmmon3 - 4*[risky]c.agec18#c.pmmon3 + 16*[risky]c.agec18#c.agec18#c.pmmon3 = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | -2.713146 .4342648 -6.25 0.000 -3.56429 -1.862003
------------------------------------------------------------------------------
. * Effect of PM Monitoring at Age 16
. lincom pmmon3*1 + c.agec18#c.pmmon3*-2 + c.agec18#c.agec18#c.pmmon3*4
( 1) [risky]pmmon3 - 2*[risky]c.agec18#c.pmmon3 + 4*[risky]c.agec18#c.agec18#c.pmmon3 = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | -3.066757 .4559046 -6.73 0.000 -3.960314 -2.173201
------------------------------------------------------------------------------
. * Effect of PM Monitoring at Age 18
. lincom pmmon3*1 + c.agec18#c.pmmon3*0 + c.agec18#c.agec18#c.pmmon3*0
( 1) [risky]pmmon3 = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | -2.590697 .5924507 -4.37 0.000 -3.751879 -1.429515
------------------------------------------------------------------------------
. predict PredPMBP, xb,
. corr risky PredPMBP
(obs=1400)
| risky PredPMBP
-------------+------------------
risky | 1.0000
PredPMBP | 0.5567 1.0000
.
. display as result "Eq 9.1: Predicting Quadratic Change in Risky Behavior"
Eq 9.1: Predicting Quadratic Change in Risky Behavior
. display as result "From Monitoring at Age 18 as Between-Person Monitoring"
From Monitoring at Age 18 as Between-Person Monitoring
. mixed risky c.agec18 c.agec18#c.agec18 c.att4 c.agec18#c.att4 ///
> c.age18mon3 c.agec18#c.age18mon3 c.agec18#c.agec18#c.age18mon3, ///
> || personid: agec18, variance mle covariance(unstructured),
Performing EM optimization:
Performing gradient-based optimization:
Iteration 0: log likelihood = -3783.7589
Iteration 1: log likelihood = -3783.7589
Computing standard errors:
Mixed-effects ML regression Number of obs = 1400
Group variable: personid Number of groups = 200
Obs per group: min = 7
avg = 7.0
max = 7
Wald chi2(7) = 446.85
Log likelihood = -3783.7589 Prob > chi2 = 0.0000
-----------------------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
------------------------------+----------------------------------------------------------------
agec18 | 1.908314 .1461234 13.06 0.000 1.621918 2.194711
|
c.agec18#c.agec18 | .1324328 .0220362 6.01 0.000 .0892427 .1756229
|
att4 | -3.324226 .5273662 -6.30 0.000 -4.357845 -2.290607
|
c.agec18#c.att4 | -.5327659 .1030131 -5.17 0.000 -.7346679 -.3308639
|
age18mon3 | -1.79405 .6036914 -2.97 0.003 -2.977263 -.6108364
|
c.agec18#c.age18mon3 | .6551895 .2606553 2.51 0.012 .1443145 1.166065
|
c.agec18#c.agec18#c.age18mon3 | .1547255 .0392713 3.94 0.000 .0777552 .2316959
|
_cons | 23.41475 .3377242 69.33 0.000 22.75282 24.07667
-----------------------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
personid: Unstructured |
var(agec18) | .4691564 .0774788 .3394268 .648469
var(_cons) | 16.16709 2.009211 12.67203 20.62611
cov(agec18,_cons) | 1.688417 .3338152 1.034151 2.342683
-----------------------------+------------------------------------------------
var(Residual) | 8.226297 .3679571 7.535823 8.980036
------------------------------------------------------------------------------
LR test vs. linear regression: chi2(3) = 623.74 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
. estat ic, n(200),
Akaike's information criterion and Bayesian information criterion
-----------------------------------------------------------------------------
Model | Obs ll(null) ll(model) df AIC BIC
-------------+---------------------------------------------------------------
. | 200 . -3783.759 12 7591.518 7631.098
-----------------------------------------------------------------------------
Note: N=200 used in calculating BIC
. * Effect of Age 18 Monitoring at Age 12
. lincom c.age18mon3*1 + c.agec18#c.age18mon3*-6 + c.agec18#c.agec18#c.age18mon3*36
( 1) [risky]age18mon3 - 6*[risky]c.agec18#c.age18mon3 + 36*[risky]c.agec18#c.agec18#c.age18mon3 = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | -.1550678 .5560915 -0.28 0.780 -1.244987 .9348515
------------------------------------------------------------------------------
. * Effect of Age 18 Monitoring at Age 14
. lincom c.age18mon3*1 + c.agec18#c.age18mon3*-4 + c.agec18#c.agec18#c.age18mon3*16
( 1) [risky]age18mon3 - 4*[risky]c.agec18#c.age18mon3 + 16*[risky]c.agec18#c.agec18#c.age18mon3 = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | -1.939199 .4518618 -4.29 0.000 -2.824832 -1.053567
------------------------------------------------------------------------------
. * Effect of Age 18 Monitoring at Age 16
. lincom c.age18mon3*1 + c.agec18#c.age18mon3*-2 + c.agec18#c.agec18#c.age18mon3*4
( 1) [risky]age18mon3 - 2*[risky]c.agec18#c.age18mon3 + 4*[risky]c.agec18#c.agec18#c.age18mon3 = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | -2.485527 .4718341 -5.27 0.000 -3.410305 -1.560749
------------------------------------------------------------------------------
. * Effect of Age 18 Monitoring at Age 18
. lincom c.age18mon3*1 + c.agec18#c.age18mon3*0 + c.agec18#c.agec18#c.age18mon3*0
( 1) [risky]age18mon3 = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | -1.79405 .6036914 -2.97 0.003 -2.977263 -.6108364
------------------------------------------------------------------------------
. predict Pred18BP, xb,
. corr risky Pred18BP
(obs=1400)
| risky Pred18BP
-------------+------------------
risky | 1.0000
Pred18BP | 0.5248 1.0000
.
. display as result "Eq 9.2: Adding Within-Person Monitoring by Quadratic Age"
Eq 9.2: Adding Within-Person Monitoring by Quadratic Age
. display as result "Using Deviation from Person Mean Monitoring as Within-Person Monitoring"
Using Deviation from Person Mean Monitoring as Within-Person Monitoring
. mixed risky c.agec18 c.agec18#c.agec18 c.att4 c.agec18#c.att4 ///
> c.pmmon3 c.agec18#c.pmmon3 c.agec18#c.agec18#c.pmmon3 ///
> c.wpmon c.agec18#c.wpmon c.agec18#c.agec18#c.wpmon, ///
> || personid: agec18, variance mle covariance(unstructured),
Performing EM optimization:
Performing gradient-based optimization:
Iteration 0: log likelihood = -3729.9756
Iteration 1: log likelihood = -3729.9756
Computing standard errors:
Mixed-effects ML regression Number of obs = 1400
Group variable: personid Number of groups = 200
Obs per group: min = 7
avg = 7.0
max = 7
Wald chi2(10) = 529.18
Log likelihood = -3729.9756 Prob > chi2 = 0.0000
--------------------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
---------------------------+----------------------------------------------------------------
agec18 | 1.920856 .1419434 13.53 0.000 1.642652 2.19906
|
c.agec18#c.agec18 | .1364754 .0208909 6.53 0.000 .0955301 .1774207
|
att4 | -3.300859 .526975 -6.26 0.000 -4.333711 -2.268007
|
c.agec18#c.att4 | -.5149291 .1101147 -4.68 0.000 -.7307499 -.2991082
|
pmmon3 | -1.730816 .6196299 -2.79 0.005 -2.945268 -.5163634
|
c.agec18#c.pmmon3 | .7578588 .2639797 2.87 0.004 .2404682 1.275249
|
c.agec18#c.agec18#c.pmmon3 | .1218731 .0394164 3.09 0.002 .0446184 .1991278
|
wpmon | 2.547224 .6084861 4.19 0.000 1.354613 3.739835
|
c.agec18#c.wpmon | -.9522503 .4559129 -2.09 0.037 -1.845823 -.0586774
|
c.agec18#c.agec18#c.wpmon | -.2158759 .0741595 -2.91 0.004 -.3612258 -.070526
|
_cons | 23.46916 .3365035 69.74 0.000 22.80962 24.12869
--------------------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
personid: Unstructured |
var(agec18) | .6108648 .091911 .4548549 .8203843
var(_cons) | 16.57912 2.02333 13.05209 21.05925
cov(agec18,_cons) | 2.134342 .3715574 1.406103 2.862581
-----------------------------+------------------------------------------------
var(Residual) | 7.341472 .3309385 6.720673 8.019616
------------------------------------------------------------------------------
LR test vs. linear regression: chi2(3) = 631.07 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
. estat ic, n(200),
Akaike's information criterion and Bayesian information criterion
-----------------------------------------------------------------------------
Model | Obs ll(null) ll(model) df AIC BIC
-------------+---------------------------------------------------------------
. | 200 . -3729.976 15 7489.951 7539.426
-----------------------------------------------------------------------------
Note: N=200 used in calculating BIC
. * Effect of PM Monitoring at Age 12
. lincom c.pmmon3*1 + c.agec18#c.pmmon3*-6 + c.agec18#c.agec18#c.pmmon3*36
( 1) [risky]pmmon3 - 6*[risky]c.agec18#c.pmmon3 + 36*[risky]c.agec18#c.agec18#c.pmmon3 = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | -1.890537 .582485 -3.25 0.001 -3.032186 -.7488869
------------------------------------------------------------------------------
. * Effect of PM Monitoring at Age 14
. lincom c.pmmon3*1 + c.agec18#c.pmmon3*-4 + c.agec18#c.agec18#c.pmmon3*16
( 1) [risky]pmmon3 - 4*[risky]c.agec18#c.pmmon3 + 16*[risky]c.agec18#c.agec18#c.pmmon3 = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | -2.812281 .4350774 -6.46 0.000 -3.665017 -1.959545
------------------------------------------------------------------------------
. * Effect of PM Monitoring at Age 16
. lincom c.pmmon3*1 + c.agec18#c.pmmon3*-2 + c.agec18#c.agec18#c.pmmon3*4
( 1) [risky]pmmon3 - 2*[risky]c.agec18#c.pmmon3 + 4*[risky]c.agec18#c.agec18#c.pmmon3 = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | -2.759041 .4578984 -6.03 0.000 -3.656505 -1.861576
------------------------------------------------------------------------------
. * Effect of PM Monitoring at Age 18
. lincom c.pmmon3*1 + c.agec18#c.pmmon3*0 + c.agec18#c.agec18#c.pmmon3*0
( 1) [risky]pmmon3 = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | -1.730816 .6196299 -2.79 0.005 -2.945268 -.5163634
------------------------------------------------------------------------------
. * Effect of WP Monitoring at Age 12
. lincom c.wpmon*1 + c.agec18#c.wpmon*-6 + c.agec18#c.agec18#c.wpmon*36
( 1) [risky]wpmon - 6*[risky]c.agec18#c.wpmon + 36*[risky]c.agec18#c.agec18#c.wpmon = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | .4891941 .6492799 0.75 0.451 -.7833712 1.761759
------------------------------------------------------------------------------
. * Effect of WP Monitoring at Age 14
. lincom c.wpmon*1 + c.agec18#c.wpmon*-4 + c.agec18#c.agec18#c.wpmon*16
( 1) [risky]wpmon - 4*[risky]c.agec18#c.wpmon + 16*[risky]c.agec18#c.agec18#c.wpmon = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | 2.902211 .3615757 8.03 0.000 2.193536 3.610886
------------------------------------------------------------------------------
. * Effect of WP Monitoring at Age 16
. lincom c.wpmon*1 + c.agec18#c.wpmon*-2 + c.agec18#c.agec18#c.wpmon*4
( 1) [risky]wpmon - 2*[risky]c.agec18#c.wpmon + 4*[risky]c.agec18#c.agec18#c.wpmon = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | 3.588221 .3693788 9.71 0.000 2.864252 4.31219
------------------------------------------------------------------------------
. * Effect of WP Monitoring at Age 18
. lincom c.wpmon*1 + c.agec18#c.wpmon*0 + c.agec18#c.agec18#c.wpmon*0
( 1) [risky]wpmon = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | 2.547224 .6084861 4.19 0.000 1.354613 3.739835
------------------------------------------------------------------------------
. predict PredPMBPWP, xb,
. corr risky PredPMBPWP
(obs=1400)
| risky PredP~WP
-------------+------------------
risky | 1.0000
PredPMBPWP | 0.5669 1.0000
.
. display as result "Eq 9.2: Adding Within-Person Monitoring by Quadratic Age"
Eq 9.2: Adding Within-Person Monitoring by Quadratic Age
. display as result "Using Change from Age 18 Monitoring as Within-Person Monitoring"
Using Change from Age 18 Monitoring as Within-Person Monitoring
. mixed risky c.agec18 c.agec18#c.agec18 c.att4 c.agec18#c.att4 ///
> c.age18mon3 c.agec18#c.age18mon3 c.agec18#c.agec18#c.age18mon3 ///
> c.change18mon c.agec18#c.change18mon c.agec18#c.agec18#c.change18mon, ///
> || personid: agec18, variance mle covariance(unstructured),
Performing EM optimization:
Performing gradient-based optimization:
Iteration 0: log likelihood = -3751.7559
Iteration 1: log likelihood = -3751.7559
Computing standard errors:
Mixed-effects ML regression Number of obs = 1400
Group variable: personid Number of groups = 200
Obs per group: min = 7
avg = 7.0
max = 7
Wald chi2(10) = 511.47
Log likelihood = -3751.7559 Prob > chi2 = 0.0000
-------------------------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
--------------------------------+----------------------------------------------------------------
agec18 | 1.94276 .1411758 13.76 0.000 1.66606 2.219459
|
c.agec18#c.agec18 | .1382142 .0210469 6.57 0.000 .0969631 .1794653
|
att4 | -3.332456 .5446091 -6.12 0.000 -4.399871 -2.265042
|
c.agec18#c.att4 | -.5361196 .1044842 -5.13 0.000 -.7409049 -.3313343
|
age18mon3 | -1.410961 .6224682 -2.27 0.023 -2.630976 -.1909458
|
c.agec18#c.age18mon3 | .3411646 .2592881 1.32 0.188 -.1670308 .8493601
|
c.agec18#c.agec18#c.age18mon3 | .0930822 .0389909 2.39 0.017 .0166613 .169503
|
change18mon | 4.716308 .8614941 5.47 0.000 3.02781 6.404805
|
c.agec18#c.change18mon | 1.198141 .4729891 2.53 0.011 .2710999 2.125183
|
c.agec18#c.agec18#c.change18mon | .0918663 .0606977 1.51 0.130 -.027099 .2108317
|
_cons | 23.4013 .3460527 67.62 0.000 22.72305 24.07955
-------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
personid: Unstructured |
var(agec18) | .5177212 .0802292 .3821101 .7014606
var(_cons) | 17.84347 2.156087 14.08075 22.61168
cov(agec18,_cons) | 1.762454 .34705 1.082249 2.44266
-----------------------------+------------------------------------------------
var(Residual) | 7.47218 .3394318 6.835662 8.167968
------------------------------------------------------------------------------
LR test vs. linear regression: chi2(3) = 666.65 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
. estat ic, n(200),
Akaike's information criterion and Bayesian information criterion
-----------------------------------------------------------------------------
Model | Obs ll(null) ll(model) df AIC BIC
-------------+---------------------------------------------------------------
. | 200 . -3751.756 15 7533.512 7582.987
-----------------------------------------------------------------------------
Note: N=200 used in calculating BIC
. * Effect of Age 18 Monitoring at Age 12
. lincom c.age18mon3*1 + c.agec18#c.age18mon3*-6 + c.agec18#c.agec18#c.age18mon3*36
( 1) [risky]age18mon3 - 6*[risky]c.agec18#c.age18mon3 + 36*[risky]c.agec18#c.agec18#c.age18mon3 = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | -.1069914 .5928511 -0.18 0.857 -1.268958 1.054976
------------------------------------------------------------------------------
. * Effect of Age 18 Monitoring at Age 14
. lincom c.age18mon3*1 + c.agec18#c.age18mon3*-4 + c.agec18#c.agec18#c.age18mon3*16
( 1) [risky]age18mon3 - 4*[risky]c.agec18#c.age18mon3 + 16*[risky]c.agec18#c.agec18#c.age18mon3 = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | -1.286305 .4895503 -2.63 0.009 -2.245806 -.3268042
------------------------------------------------------------------------------
. * Effect of Age 18 Monitoring at Age 16
. lincom c.age18mon3*1 + c.agec18#c.age18mon3*-2 + c.agec18#c.agec18#c.age18mon3*4
( 1) [risky]age18mon3 - 2*[risky]c.agec18#c.age18mon3 + 4*[risky]c.agec18#c.agec18#c.age18mon3 = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | -1.720962 .5037073 -3.42 0.001 -2.70821 -.7337136
------------------------------------------------------------------------------
. * Effect of Age 18 Monitoring at Age 18
. lincom c.age18mon3*1 + c.agec18#c.age18mon3*0 + c.agec18#c.agec18#c.age18mon3*0
( 1) [risky]age18mon3 = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | -1.410961 .6224682 -2.27 0.023 -2.630976 -.1909458
------------------------------------------------------------------------------
. * Effect of Change in Monitoring at Age 12
. lincom c.change18mon*1 + c.agec18#c.change18mon*-6 + c.agec18#c.agec18#c.change18mon*36
( 1) [risky]change18mon - 6*[risky]c.agec18#c.change18mon + 36*[risky]c.agec18#c.agec18#c.change18mon = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | .8346474 .3722552 2.24 0.025 .1050405 1.564254
------------------------------------------------------------------------------
. * Effect of Change in Monitoring at Age 14
. lincom c.change18mon*1 + c.agec18#c.change18mon*-4 + c.agec18#c.agec18#c.change18mon*16
( 1) [risky]change18mon - 4*[risky]c.agec18#c.change18mon + 16*[risky]c.agec18#c.agec18#c.change18mon = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | 1.393603 .297395 4.69 0.000 .81072 1.976487
------------------------------------------------------------------------------
. * Effect of Change in Monitoring at Age 16
. lincom c.change18mon*1 + c.agec18#c.change18mon*-2 + c.agec18#c.agec18#c.change18mon*4
( 1) [risky]change18mon - 2*[risky]c.agec18#c.change18mon + 4*[risky]c.agec18#c.agec18#c.change18mon = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | 2.68749 .3234863 8.31 0.000 2.053469 3.321512
------------------------------------------------------------------------------
. * Effect of Change in Monitoring at Age 18
. lincom c.change18mon*1 + c.agec18#c.change18mon*0 + c.agec18#c.agec18#c.change18mon*0
( 1) [risky]change18mon = 0
------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | 4.716308 .8614941 5.47 0.000 3.02781 6.404805
------------------------------------------------------------------------------
. predict Pred18BPWP, xb,
. corr risky Pred18BPWP
(obs=1400)
| risky Pred1~WP
-------------+------------------
risky | 1.0000
Pred18BPWP | 0.4894 1.0000
. display as result "Ch 9: Random Linear Age Model for Monitoring"
Ch 9: Random Linear Age Model for Monitoring
. display as result "Saving Predicted Random Effects and Residuals as Data"
Saving Predicted Random Effects and Residuals as Data
. mixed monitor c.agec18, ///
> || personid: agec18, variance mle covariance(unstructured),
Performing EM optimization:
Performing gradient-based optimization:
Iteration 0: log likelihood = -676.87612
Iteration 1: log likelihood = -676.87612
Computing standard errors:
Mixed-effects ML regression Number of obs = 1400
Group variable: personid Number of groups = 200
Obs per group: min = 7
avg = 7.0
max = 7
Wald chi2(1) = 0.16
Log likelihood = -676.87612 Prob > chi2 = 0.6861
------------------------------------------------------------------------------
monitor | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
agec18 | -.0033057 .0081793 -0.40 0.686 -.0193369 .0127255
_cons | 3.065015 .034128 89.81 0.000 2.998126 3.131905
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
personid: Unstructured |
var(agec18) | .0104944 .0013447 .0081638 .0134905
var(_cons) | .1954436 .0233316 .1546704 .2469652
cov(agec18,_cons) | -.000423 .0040081 -.0082787 .0074326
-----------------------------+------------------------------------------------
var(Residual) | .0807506 .0036117 .0739732 .0881489
------------------------------------------------------------------------------
LR test vs. linear regression: chi2(3) = 1388.33 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
. estat ic, n(200),
Akaike's information criterion and Bayesian information criterion
-----------------------------------------------------------------------------
Model | Obs ll(null) ll(model) df AIC BIC
-------------+---------------------------------------------------------------
. | 200 . -676.8761 6 1365.752 1385.542
-----------------------------------------------------------------------------
Note: N=200 used in calculating BIC
. predict monUage monU0, reffects,
. predict monEres, residuals
.
. * Center random intercept at 3 (have to add fixed effect)
. gen monUint = monU0+3.0650-3
.
. display as result "Descriptives for Random Effects and Residuals"
Descriptives for Random Effects and Residuals
. summarize(monUint monUage monEres)
Variable | Obs Mean Std. Dev. Min Max
-------------+--------------------------------------------------------
monUint | 1400 .065 .4101274 -1.497594 1.149218
monUage | 1400 -1.81e-10 .091895 -.2221221 .2739341
monEres | 1400 3.42e-10 .2482451 -.7779043 .757265
.
. display as result "Ch 9 Eq 9.6: Predicting Risky Behavior"
Ch 9 Eq 9.6: Predicting Risky Behavior
. display as result "from Monitoring Random Effects and Residuals"
from Monitoring Random Effects and Residuals
. mixed risky c.agec18 c.agec18#c.agec18 c.att4 c.agec18#c.att4 ///
> c.monUint c.monUint#c.agec18 c.monUage c.monUage#c.agec18 c.monEres, ///
> || personid: agec18, variance mle covariance(unstructured),
Performing EM optimization:
Performing gradient-based optimization:
Iteration 0: log likelihood = -3715.4007
Iteration 1: log likelihood = -3715.4007
Computing standard errors:
Mixed-effects ML regression Number of obs = 1400
Group variable: personid Number of groups = 200
Obs per group: min = 7
avg = 7.0
max = 7
Wald chi2(9) = 602.72
Log likelihood = -3715.4007 Prob > chi2 = 0.0000
------------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------------+----------------------------------------------------------------
agec18 | 2.00976 .1384176 14.52 0.000 1.738467 2.281054
|
c.agec18#c.agec18 | .1465475 .0205815 7.12 0.000 .1062086 .1868865
|
att4 | -3.331767 .5137492 -6.49 0.000 -4.338697 -2.324837
|
c.agec18#c.att4 | -.529372 .1027295 -5.15 0.000 -.7307182 -.3280258
|
monUint | -4.359074 .7652822 -5.70 0.000 -5.859 -2.859149
|
c.monUint#c.agec18 | -.5474589 .1530265 -3.58 0.000 -.8473854 -.2475325
|
monUage | 3.756083 3.412991 1.10 0.271 -2.933257 10.44542
|
c.monUage#c.agec18 | -1.748057 .6858482 -2.55 0.011 -3.092294 -.403819
|
monEres | 3.557998 .3013236 11.81 0.000 2.967414 4.148581
_cons | 23.59709 .3294869 71.62 0.000 22.95131 24.24287
------------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
personid: Unstructured |
var(agec18) | .498307 .0769795 .3681303 .6745161
var(_cons) | 15.59704 1.908355 12.27141 19.82393
cov(agec18,_cons) | 1.799114 .32767 1.156892 2.441335
-----------------------------+------------------------------------------------
var(Residual) | 7.331196 .3278895 6.715906 8.002856
------------------------------------------------------------------------------
LR test vs. linear regression: chi2(3) = 635.17 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
. estat ic, n(200),
Akaike's information criterion and Bayesian information criterion
-----------------------------------------------------------------------------
Model | Obs ll(null) ll(model) df AIC BIC
-------------+---------------------------------------------------------------
. | 200 . -3715.401 14 7458.801 7504.978
-----------------------------------------------------------------------------
Note: N=200 used in calculating BIC
. estimates store FitUWPasfixed,
.
. display as result "Ch 9 Eq 9.6: Predicting Risky Behavior"
Ch 9 Eq 9.6: Predicting Risky Behavior
. display as result "from Monitoring Random Effects and Residuals"
from Monitoring Random Effects and Residuals
. display as result "Adding Random Effect of WP Monitoring Residual"
Adding Random Effect of WP Monitoring Residual
. mixed risky c.agec18 c.agec18#c.agec18 c.att4 c.agec18#c.att4 ///
> c.monUint c.monUint#c.agec18 c.monUage c.monUage#c.agec18 c.monEres, ///
> || personid: agec18 monEres, variance mle covariance(unstructured),
Performing EM optimization:
Performing gradient-based optimization:
Iteration 0: log likelihood = -3712.9942
Iteration 1: log likelihood = -3710.6298
Iteration 2: log likelihood = -3710.4508
Iteration 3: log likelihood = -3710.4025
Iteration 4: log likelihood = -3710.3879
Iteration 5: log likelihood = -3710.3854
Iteration 6: log likelihood = -3710.3853
Computing standard errors:
Mixed-effects ML regression Number of obs = 1400
Group variable: personid Number of groups = 200
Obs per group: min = 7
avg = 7.0
max = 7
Wald chi2(9) = 589.48
Log likelihood = -3710.3853 Prob > chi2 = 0.0000
------------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------------+----------------------------------------------------------------
agec18 | 2.009853 .1378964 14.58 0.000 1.739581 2.280125
|
c.agec18#c.agec18 | .146499 .0204683 7.16 0.000 .1063819 .1866162
|
att4 | -3.205897 .5052471 -6.35 0.000 -4.196163 -2.215631
|
c.agec18#c.att4 | -.5261396 .1029355 -5.11 0.000 -.7278895 -.3243897
|
monUint | -4.391419 .7537818 -5.83 0.000 -5.868804 -2.914033
|
c.monUint#c.agec18 | -.5520106 .1533508 -3.60 0.000 -.8525727 -.2514485
|
monUage | 3.619208 3.373788 1.07 0.283 -2.993296 10.23171
|
c.monUage#c.agec18 | -1.754342 .6874393 -2.55 0.011 -3.101698 -.4069858
|
monEres | 3.517566 .3083171 11.41 0.000 2.913276 4.121857
_cons | 23.60472 .3280723 71.95 0.000 22.96171 24.24773
------------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
personid: Unstructured |
var(agec18) | .5046969 . . .
var(monEres) | 1.030301 . . .
var(_cons) | 15.5873 . . .
cov(agec18,monEres) | .0368036 . . .
cov(agec18,_cons) | 1.803278 . . .
cov(monEres,_cons) | 3.196908 . . .
-----------------------------+------------------------------------------------
var(Residual) | 7.240263 . . .
------------------------------------------------------------------------------
LR test vs. linear regression: chi2(6) = 645.20 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
. estat ic, n(200),
Akaike's information criterion and Bayesian information criterion
-----------------------------------------------------------------------------
Model | Obs ll(null) ll(model) df AIC BIC
-------------+---------------------------------------------------------------
. | 200 . -3710.385 10 7440.771 7473.754
-----------------------------------------------------------------------------
Note: N=200 used in calculating BIC
. estimates store FitUWPasrandom,
. lrtest FitUWPasrandom FitUWPasfixed, force
Likelihood-ratio test LR chi2(4) = -10.03
(Assumption: FitUWPasrandom nested in FitUWPasfixed) Prob > chi2 = 1.0000
.
. display as result "Ch 9 Eq 9.6: Predicting Risky Behavior"
Ch 9 Eq 9.6: Predicting Risky Behavior
. display as result "from Monitoring Random Effects and Residuals"
from Monitoring Random Effects and Residuals
. display as result "Adding WP Monitoring by Age"
Adding WP Monitoring by Age
. mixed risky c.agec18 c.agec18#c.agec18 c.att4 c.agec18#c.att4 ///
> c.monUint c.monUint#c.agec18 c.monUage c.monUage#c.agec18 ///
> c.monEres c.agec18#c.monEres, ///
> || personid: agec18, variance mle covariance(unstructured),
Performing EM optimization:
Performing gradient-based optimization:
Iteration 0: log likelihood = -3710.5794
Iteration 1: log likelihood = -3710.5794
Computing standard errors:
Mixed-effects ML regression Number of obs = 1400
Group variable: personid Number of groups = 200
Obs per group: min = 7
avg = 7.0
max = 7
Wald chi2(10) = 614.32
Log likelihood = -3710.5794 Prob > chi2 = 0.0000
------------------------------------------------------------------------------------
risky | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------------+----------------------------------------------------------------
agec18 | 2.006329 .1378934 14.55 0.000 1.736063 2.276595
|
c.agec18#c.agec18 | .1452149 .0204877 7.09 0.000 .1050599 .18537
|
att4 | -3.306071 .5137122 -6.44 0.000 -4.312929 -2.299214
|
c.agec18#c.att4 | -.520759 .1027575 -5.07 0.000 -.72216 -.319358
|
monUint | -4.615041 .7695191 -6.00 0.000 -6.123271 -3.106812
|
c.monUint#c.agec18 | -.6000601 .1539402 -3.90 0.000 -.9017773 -.2983428
|
monUage | 3.060632 3.419524 0.90 0.371 -3.641513 9.762776
|
c.monUage#c.agec18 | -1.778625 .685806 -2.59 0.010 -3.12278 -.4344703
|
monEres | 5.201006 .6070731 8.57 0.000 4.011164 6.390847
|
c.agec18#c.monEres | .5444871 .1749224 3.11 0.002 .2016455 .8873287
|
_cons | 23.61048 .3293018 71.70 0.000 22.96506 24.2559
------------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
personid: Unstructured |
var(agec18) | .5006706 .0769436 .3704576 .6766524
var(_cons) | 15.62195 1.907302 12.29734 19.84538
cov(agec18,_cons) | 1.805091 .3274413 1.163318 2.446864
-----------------------------+------------------------------------------------
var(Residual) | 7.26096 .3247509 6.65156 7.926191
------------------------------------------------------------------------------
LR test vs. linear regression: chi2(3) = 640.97 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
. estat ic, n(200),
Akaike's information criterion and Bayesian information criterion
-----------------------------------------------------------------------------
Model | Obs ll(null) ll(model) df AIC BIC
-------------+---------------------------------------------------------------
. | 200 . -3710.579 15 7451.159 7500.634
-----------------------------------------------------------------------------
Note: N=200 used in calculating BIC
.
. ****** END CHAPTER 9 MODELS ******
.
. * Close log
. log close STATA_Chapter9
name: STATA_Chapter9
log: C:\Dropbox\PilesOfVariance\Chapter9\STATA\STATA_Chapter9_Output.smcl
log type: smcl
closed on: 20 Jan 2015, 16:07:57
------------------------------------------------------------------------------------------------------------------------------------------------------