use https://www.bristol.ac.uk/cmm/media/runmlwin/tutorial, clearreffadjust4nlcom helpfile examples
Examples 1 & 2 assume the path to the MLwiN executable is set in global MLwiN_path; see help runmlwin.
Example 1: Two level continuous response model
(See page 59 of the MLwiN User Manual)
- read in data
- fit model using MLwiN via
runmlwin
runmlwin normexam cons standlrt, level1(student: cons) level2(school:cons standlrt) batch
--- Begin MLwiN error log ---
MLN - Software for N-level analysis. Mon 20 Oct 2025 08:53:58
/var/folders/1l/w0qjrymd3hb44s3skrzn4w4w0000gn/T//St94012.000003
/var/folders/1l/w0qjrymd3hb44s3skrzn4w4w0000gn/T//St94012.000007
--- End MLwiN error log ---
Warning: getversion plugin could not be loaded
MLwiN 3.16 multilevel model Number of obs = 4059
Normal response model (hierarchical)
Estimation algorithm: IGLS
-----------------------------------------------------------
| No. of Observations per Group
Level Variable | Groups Minimum Average Maximum
----------------+------------------------------------------
school | 65 2 62.4 198
-----------------------------------------------------------
Run time (seconds) = 0.08
Number of iterations = 4
Log likelihood = -4658.435
Deviance = 9316.87
------------------------------------------------------------------------------
normexam | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
cons | -.0115051 .039783 -0.29 0.772 -.0894783 .066468
standlrt | .5567305 .019937 27.92 0.000 .5176547 .5958062
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
Level 2: school |
var(cons) | .0904446 .017924 .0553142 .1255749
cov(cons,standlrt) | .0180414 .0067229 .0048649 .031218
var(standlrt) | .0145361 .0044139 .0058851 .0231872
-----------------------------+------------------------------------------------
Level 1: student |
var(cons) | .5536575 .0124818 .5291937 .5781214
------------------------------------------------------------------------------/nogui option ignored
ECHO 0
Execution completed- report coefficient and delta-method confidence interval
reffadjust4nlcom cons standlrt, eqn(RP2)
nlcom `r(beta_standlrt)'
_nl_1: [RP2]cov(cons\standlrt)/[RP2]var(standlrt)
------------------------------------------------------------------------------
| Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
_nl_1 | 1.241144 .4275471 2.90 0.004 .4031675 2.079121
------------------------------------------------------------------------------- compare reporting coefficient as median with 2.5 & 97.5 percentiles
reffadjustsim cons standlrt, eqn(RP2) seed(12345)
------------------------------------------------------------------------------
cons | Median 2.5 Percentile 97.5 Percentile
-------------+----------------------------------------------------------------
standlrt | 1.242541 .4311653 2.607229
------------------------------------------------------------------------------- compare reporting coefficient as mean & Wald-type confidence interval
- Warning: mean and Wald-type confidence are inaccurate in this example
reffadjustsim cons standlrt, eqn(RP2) seed(12345) waldtype
------------------------------------------------------------------------------
cons | Coef. Std. Err. [95% Conf. Interval]
-------------+----------------------------------------------------------------
standlrt | 1.319134 1.22225 -1.076431 3.7147
------------------------------------------------------------------------------
Warning: Coef. & Wald-type conf. interval limits may be inaccurate.
Please compare with default output which reports median & centiles.- to view just the coefficient or string expression for the coefficient
reffadjust4nlcom cons standlrt, eqn(RP2)
display `r(beta_standlrt)'
mata st_macroexpand("`r(beta_standlrt)'")1.2411444
[RP2]cov(cons\standlrt)/[RP2]var(standlrt)- compare with Bayesian posterior distribution
runmlwin normexam cons standlrt, level1(student: cons) level2(school:cons standlrt) batch mcmc(on) initsprevious seed(121211)
mcmcsum, getchains
reffadjust4nlcom cons standlrt, eqn(RP2) mcmcsum
gen beta_standlrt = `r(beta_standlrt)'
mcmcsum beta_standlrt, variables /nogui option ignored
ECHO 0
BURNING IN...
Burning in for 50 iterations out of 500
Burning in for 100 iterations out of 500
Burning in for 150 iterations out of 500
Burning in for 200 iterations out of 500
Burning in for 250 iterations out of 500
Burning in for 300 iterations out of 500
Burning in for 350 iterations out of 500
Burning in for 400 iterations out of 500
Burning in for 450 iterations out of 500
Burning in for 500 iterations out of 500
Actual update 50 of 5000, Stored update 50 of 5000
Actual update 100 of 5000, Stored update 100 of 5000
Actual update 150 of 5000, Stored update 150 of 5000
Actual update 200 of 5000, Stored update 200 of 5000
Actual update 250 of 5000, Stored update 250 of 5000
Actual update 300 of 5000, Stored update 300 of 5000
Actual update 350 of 5000, Stored update 350 of 5000
Actual update 400 of 5000, Stored update 400 of 5000
Actual update 450 of 5000, Stored update 450 of 5000
Actual update 500 of 5000, Stored update 500 of 5000
Actual update 550 of 5000, Stored update 550 of 5000
Actual update 600 of 5000, Stored update 600 of 5000
Actual update 650 of 5000, Stored update 650 of 5000
Actual update 700 of 5000, Stored update 700 of 5000
Actual update 750 of 5000, Stored update 750 of 5000
Actual update 800 of 5000, Stored update 800 of 5000
Actual update 850 of 5000, Stored update 850 of 5000
Actual update 900 of 5000, Stored update 900 of 5000
Actual update 950 of 5000, Stored update 950 of 5000
Actual update 1000 of 5000, Stored update 1000 of 5000
Actual update 1050 of 5000, Stored update 1050 of 5000
Actual update 1100 of 5000, Stored update 1100 of 5000
Actual update 1150 of 5000, Stored update 1150 of 5000
Actual update 1200 of 5000, Stored update 1200 of 5000
Actual update 1250 of 5000, Stored update 1250 of 5000
Actual update 1300 of 5000, Stored update 1300 of 5000
Actual update 1350 of 5000, Stored update 1350 of 5000
Actual update 1400 of 5000, Stored update 1400 of 5000
Actual update 1450 of 5000, Stored update 1450 of 5000
Actual update 1500 of 5000, Stored update 1500 of 5000
Actual update 1550 of 5000, Stored update 1550 of 5000
Actual update 1600 of 5000, Stored update 1600 of 5000
Actual update 1650 of 5000, Stored update 1650 of 5000
Actual update 1700 of 5000, Stored update 1700 of 5000
Actual update 1750 of 5000, Stored update 1750 of 5000
Actual update 1800 of 5000, Stored update 1800 of 5000
Actual update 1850 of 5000, Stored update 1850 of 5000
Actual update 1900 of 5000, Stored update 1900 of 5000
Actual update 1950 of 5000, Stored update 1950 of 5000
Actual update 2000 of 5000, Stored update 2000 of 5000
Actual update 2050 of 5000, Stored update 2050 of 5000
Actual update 2100 of 5000, Stored update 2100 of 5000
Actual update 2150 of 5000, Stored update 2150 of 5000
Actual update 2200 of 5000, Stored update 2200 of 5000
Actual update 2250 of 5000, Stored update 2250 of 5000
Actual update 2300 of 5000, Stored update 2300 of 5000
Actual update 2350 of 5000, Stored update 2350 of 5000
Actual update 2400 of 5000, Stored update 2400 of 5000
Actual update 2450 of 5000, Stored update 2450 of 5000
Actual update 2500 of 5000, Stored update 2500 of 5000
Actual update 2550 of 5000, Stored update 2550 of 5000
Actual update 2600 of 5000, Stored update 2600 of 5000
Actual update 2650 of 5000, Stored update 2650 of 5000
Actual update 2700 of 5000, Stored update 2700 of 5000
Actual update 2750 of 5000, Stored update 2750 of 5000
Actual update 2800 of 5000, Stored update 2800 of 5000
Actual update 2850 of 5000, Stored update 2850 of 5000
Actual update 2900 of 5000, Stored update 2900 of 5000
Actual update 2950 of 5000, Stored update 2950 of 5000
Actual update 3000 of 5000, Stored update 3000 of 5000
Actual update 3050 of 5000, Stored update 3050 of 5000
Actual update 3100 of 5000, Stored update 3100 of 5000
Actual update 3150 of 5000, Stored update 3150 of 5000
Actual update 3200 of 5000, Stored update 3200 of 5000
Actual update 3250 of 5000, Stored update 3250 of 5000
Actual update 3300 of 5000, Stored update 3300 of 5000
Actual update 3350 of 5000, Stored update 3350 of 5000
Actual update 3400 of 5000, Stored update 3400 of 5000
Actual update 3450 of 5000, Stored update 3450 of 5000
Actual update 3500 of 5000, Stored update 3500 of 5000
Actual update 3550 of 5000, Stored update 3550 of 5000
Actual update 3600 of 5000, Stored update 3600 of 5000
Actual update 3650 of 5000, Stored update 3650 of 5000
Actual update 3700 of 5000, Stored update 3700 of 5000
Actual update 3750 of 5000, Stored update 3750 of 5000
Actual update 3800 of 5000, Stored update 3800 of 5000
Actual update 3850 of 5000, Stored update 3850 of 5000
Actual update 3900 of 5000, Stored update 3900 of 5000
Actual update 3950 of 5000, Stored update 3950 of 5000
Actual update 4000 of 5000, Stored update 4000 of 5000
Actual update 4050 of 5000, Stored update 4050 of 5000
Actual update 4100 of 5000, Stored update 4100 of 5000
Actual update 4150 of 5000, Stored update 4150 of 5000
Actual update 4200 of 5000, Stored update 4200 of 5000
Actual update 4250 of 5000, Stored update 4250 of 5000
Actual update 4300 of 5000, Stored update 4300 of 5000
Actual update 4350 of 5000, Stored update 4350 of 5000
Actual update 4400 of 5000, Stored update 4400 of 5000
Actual update 4450 of 5000, Stored update 4450 of 5000
Actual update 4500 of 5000, Stored update 4500 of 5000 --- Begin MLwiN error log ---
MLN - Software for N-level analysis. Mon 20 Oct 2025 08:53:58
/var/folders/1l/w0qjrymd3hb44s3skrzn4w4w0000gn/T//St94012.000007
/var/folders/1l/w0qjrymd3hb44s3skrzn4w4w0000gn/T//St94012.00000b
/var/folders/1l/w0qjrymd3hb44s3skrzn4w4w0000gn/T//St94012.000005
--- End MLwiN error log ---
Warning: getversion plugin could not be loaded
MLwiN 3.16 multilevel model Number of obs = 4059
Normal response model (hierarchical)
Estimation algorithm: MCMC
-----------------------------------------------------------
| No. of Observations per Group
Level Variable | Groups Minimum Average Maximum
----------------+------------------------------------------
school | 65 2 62.4 198
-----------------------------------------------------------
Burnin = 500
Chain = 5000
Thinning = 1
Run time (seconds) = .613
Deviance (dbar) = 9122.42
Deviance (thetabar) = 9031.09
Effective no. of pars (pd) = 91.34
Bayesian DIC = 9213.76
------------------------------------------------------------------------------
normexam | Mean Std. Dev. ESS P [95% Cred. Interval]
-------------+----------------------------------------------------------------
cons | -.010594 .040224 189 0.395 -.0908355 .0678323
standlrt | .5568075 .0200344 839 0.000 .5167302 .5960792
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Mean Std. Dev. ESS [95% Cred. Int]
-----------------------------+------------------------------------------------
Level 2: school |
var(cons) | .0961163 .0199991 2715 .0644542 .1421674
cov(cons,standlrt) | .0190683 .0072205 1713 .0065699 .034989
var(standlrt) | .0153485 .0046893 978 .0078805 .0261361
-----------------------------+------------------------------------------------
Level 1: student |
var(cons) | .5542075 .0123209 4645 .5309043 .578929
------------------------------------------------------------------------------Actual update 4550 of 5000, Stored update 4550 of 5000
Actual update 4600 of 5000, Stored update 4600 of 5000
Actual update 4650 of 5000, Stored update 4650 of 5000
Actual update 4700 of 5000, Stored update 4700 of 5000
Actual update 4750 of 5000, Stored update 4750 of 5000
Actual update 4800 of 5000, Stored update 4800 of 5000
Actual update 4850 of 5000, Stored update 4850 of 5000
Actual update 4900 of 5000, Stored update 4900 of 5000
Actual update 4950 of 5000, Stored update 4950 of 5000
Actual update 5000 of 5000, Stored update 5000 of 5000
Execution completed------------------------------------------------------------------------------
| Mean Std. Dev. ESS P [95% Cred. Interval]
-------------+----------------------------------------------------------------
beta_standlrt| 1.289067 .456183 857 0.001 .4577979 2.279
------------------------------------------------------------------------------Example 2: Multivariate response model
(See page 214 of the MLwiN User Manual)
- read in data
use https://www.bristol.ac.uk/cmm/media/runmlwin/gcsemv1, clear- fit model using MLwiN via
runmlwin
runmlwin (written cons female, eq(1)) (csework cons female, eq(2)), ///
level1(student: (cons, eq(1)) (cons, eq(2))) ///
level2(school: (cons, eq(1)) (cons, eq(2))) ///
batch
--- Begin MLwiN error log ---
MLN - Software for N-level analysis. Mon 20 Oct 2025 08:54:00
/var/folders/1l/w0qjrymd3hb44s3skrzn4w4w0000gn/T//St94012.000006
/var/folders/1l/w0qjrymd3hb44s3skrzn4w4w0000gn/T//St94012.00000a
/var/folders/1l/w0qjrymd3hb44s3skrzn4w4w0000gn/T//St94012.00000b
--- End MLwiN error log ---
Warning: getversion plugin could not be loaded
MLwiN 3.16 multilevel model Number of obs = 1905
Multivariate response model (hierarchical)
Estimation algorithm: IGLS
-----------------------------------------------------------
| No. of Observations per Group
Level Variable | Groups Minimum Average Maximum
----------------+------------------------------------------
school | 73 2 26.1 104
-----------------------------------------------------------
Run time (seconds) = 0.13
Number of iterations = 4
Log likelihood = -13400.244
Deviance = 26800.488
------------------------------------------------------------------------------
| Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
written |
cons_1 | 49.45213 .9338433 52.96 0.000 47.62183 51.28243
female_1 | -2.50295 .5607219 -4.46 0.000 -3.601945 -1.403955
-------------+----------------------------------------------------------------
csework |
cons_2 | 69.67166 1.171786 59.46 0.000 67.375 71.96831
female_2 | 6.751393 .6706493 10.07 0.000 5.436944 8.065841
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
Level 2: school |
var(cons_1) | 46.81298 9.18733 28.80615 64.81982
cov(cons_1,cons_2) | 24.87783 8.880358 7.47265 42.28301
var(cons_2) | 75.16623 14.56485 46.61965 103.7128
-----------------------------+------------------------------------------------
Level 1: student |
var(cons_1) | 124.6343 4.349834 116.1088 133.1598
cov(cons_1,cons_2) | 73.00323 4.17829 64.81393 81.19252
var(cons_2) | 180.0982 6.245801 167.8566 192.3397
------------------------------------------------------------------------------/nogui option ignored
ECHO 0
Execution completed- report coefficient and delta-method confidence interval
reffadjust4nlcom cons_1 cons_2, eqn(RP2)
nlcom `r(beta_cons_2)'
_nl_1: [RP2]cov(cons_1\cons_2)/[RP2]var(cons_2)
------------------------------------------------------------------------------
| Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
_nl_1 | .3309709 .0986417 3.36 0.001 .1376367 .5243051
------------------------------------------------------------------------------- compare reporting coefficient as median with 2.5 and 97.5 percentiles
reffadjustsim cons_1 cons_2, eqn(RP2) seed(12345)
------------------------------------------------------------------------------
cons_1 | Median 2.5 Percentile 97.5 Percentile
-------------+----------------------------------------------------------------
cons_2 | .3308853 .1231026 .5395364
------------------------------------------------------------------------------- compare reporting coefficient as mean with Wald-type confidence interval
reffadjustsim cons_1 cons_2, eqn(RP2) seed(12345) waldtype
------------------------------------------------------------------------------
cons_1 | Coef. Std. Err. [95% Conf. Interval]
-------------+----------------------------------------------------------------
cons_2 | .330939 .1065639 .1220777 .5398003
------------------------------------------------------------------------------
Warning: Coef. & Wald-type conf. interval limits may be inaccurate.
Please compare with default output which reports median & centiles.- to view just the coefficient or string expression for the coefficient
reffadjust4nlcom cons_1 cons_2, eqn(RP2)
display `r(beta_cons_2)'
mata st_macroexpand("`r(beta_cons_2)'").33097086
[RP2]cov(cons_1\cons_2)/[RP2]var(cons_2)- compare with Bayesian posterior distribution
runmlwin (written cons female, eq(1)) (csework cons female, eq(2)), ///
level1(student: (cons, eq(1)) (cons, eq(2))) ///
level2(school: (cons, eq(1)) (cons, eq(2))) ///
batch mcmc(on) initsprevious seed(121211)
mcmcsum, getchains
reffadjust4nlcom cons_1 cons_2, eqn(RP2) mcmcsum
gen beta_cons_2 = `r(beta_cons_2)'
mcmcsum beta_cons_2, variables /nogui option ignored
ECHO 0
BURNING IN...
Burning in for 50 iterations out of 500
Burning in for 100 iterations out of 500
Burning in for 150 iterations out of 500
Burning in for 200 iterations out of 500
Burning in for 250 iterations out of 500
Burning in for 300 iterations out of 500
Burning in for 350 iterations out of 500
Burning in for 400 iterations out of 500
Burning in for 450 iterations out of 500
Burning in for 500 iterations out of 500
Actual update 50 of 5000, Stored update 50 of 5000
Actual update 100 of 5000, Stored update 100 of 5000
Actual update 150 of 5000, Stored update 150 of 5000
Actual update 200 of 5000, Stored update 200 of 5000
Actual update 250 of 5000, Stored update 250 of 5000
Actual update 300 of 5000, Stored update 300 of 5000
Actual update 350 of 5000, Stored update 350 of 5000
Actual update 400 of 5000, Stored update 400 of 5000
Actual update 450 of 5000, Stored update 450 of 5000
Actual update 500 of 5000, Stored update 500 of 5000
Actual update 550 of 5000, Stored update 550 of 5000
Actual update 600 of 5000, Stored update 600 of 5000
Actual update 650 of 5000, Stored update 650 of 5000
Actual update 700 of 5000, Stored update 700 of 5000
Actual update 750 of 5000, Stored update 750 of 5000
Actual update 800 of 5000, Stored update 800 of 5000
Actual update 850 of 5000, Stored update 850 of 5000
Actual update 900 of 5000, Stored update 900 of 5000
Actual update 950 of 5000, Stored update 950 of 5000
Actual update 1000 of 5000, Stored update 1000 of 5000
Actual update 1050 of 5000, Stored update 1050 of 5000
Actual update 1100 of 5000, Stored update 1100 of 5000
Actual update 1150 of 5000, Stored update 1150 of 5000
Actual update 1200 of 5000, Stored update 1200 of 5000
Actual update 1250 of 5000, Stored update 1250 of 5000
Actual update 1300 of 5000, Stored update 1300 of 5000
Actual update 1350 of 5000, Stored update 1350 of 5000
Actual update 1400 of 5000, Stored update 1400 of 5000
Actual update 1450 of 5000, Stored update 1450 of 5000
Actual update 1500 of 5000, Stored update 1500 of 5000
Actual update 1550 of 5000, Stored update 1550 of 5000
Actual update 1600 of 5000, Stored update 1600 of 5000
Actual update 1650 of 5000, Stored update 1650 of 5000
Actual update 1700 of 5000, Stored update 1700 of 5000
Actual update 1750 of 5000, Stored update 1750 of 5000
Actual update 1800 of 5000, Stored update 1800 of 5000
Actual update 1850 of 5000, Stored update 1850 of 5000
Actual update 1900 of 5000, Stored update 1900 of 5000
Actual update 1950 of 5000, Stored update 1950 of 5000
Actual update 2000 of 5000, Stored update 2000 of 5000
Actual update 2050 of 5000, Stored update 2050 of 5000
Actual update 2100 of 5000, Stored update 2100 of 5000
Actual update 2150 of 5000, Stored update 2150 of 5000
Actual update 2200 of 5000, Stored update 2200 of 5000
Actual update 2250 of 5000, Stored update 2250 of 5000
Actual update 2300 of 5000, Stored update 2300 of 5000
Actual update 2350 of 5000, Stored update 2350 of 5000
Actual update 2400 of 5000, Stored update 2400 of 5000
Actual update 2450 of 5000, Stored update 2450 of 5000
Actual update 2500 of 5000, Stored update 2500 of 5000
Actual update 2550 of 5000, Stored update 2550 of 5000
Actual update 2600 of 5000, Stored update 2600 of 5000
Actual update 2650 of 5000, Stored update 2650 of 5000
Actual update 2700 of 5000, Stored update 2700 of 5000
Actual update 2750 of 5000, Stored update 2750 of 5000
Actual update 2800 of 5000, Stored update 2800 of 5000
Actual update 2850 of 5000, Stored update 2850 of 5000
Actual update 2900 of 5000, Stored update 2900 of 5000
Actual update 2950 of 5000, Stored update 2950 of 5000
Actual update 3000 of 5000, Stored update 3000 of 5000
Actual update 3050 of 5000, Stored update 3050 of 5000
Actual update 3100 of 5000, Stored update 3100 of 5000
Actual update 3150 of 5000, Stored update 3150 of 5000
Actual update 3200 of 5000, Stored update 3200 of 5000
Actual update 3250 of 5000, Stored update 3250 of 5000
Actual update 3300 of 5000, Stored update 3300 of 5000
Actual update 3350 of 5000, Stored update 3350 of 5000
Actual update 3400 of 5000, Stored update 3400 of 5000
Actual update 3450 of 5000, Stored update 3450 of 5000
Actual update 3500 of 5000, Stored update 3500 of 5000
Actual update 3550 of 5000, Stored update 3550 of 5000
Actual update 3600 of 5000, Stored update 3600 of 5000
Actual update 3650 of 5000, Stored update 3650 of 5000
Actual update 3700 of 5000, Stored update 3700 of 5000
Actual update 3750 of 5000, Stored update 3750 of 5000
Actual update 3800 of 5000, Stored update 3800 of 5000
Actual update 3850 of 5000, Stored update 3850 of 5000
Actual update 3900 of 5000, Stored update 3900 of 5000
Actual update 3950 of 5000, Stored update 3950 of 5000
Actual update 4000 of 5000, Stored update 4000 of 5000
Actual update 4050 of 5000, Stored update 4050 of 5000
Actual update 4100 of 5000, Stored update 4100 of 5000
Actual update 4150 of 5000, Stored update 4150 of 5000
Actual update 4200 of 5000, Stored update 4200 of 5000
Actual update 4250 of 5000, Stored update 4250 of 5000
Actual update 4300 of 5000, Stored update 4300 of 5000
Actual update 4350 of 5000, Stored update 4350 of 5000
Actual update 4400 of 5000, Stored update 4400 of 5000
Actual update 4450 of 5000, Stored update 4450 of 5000
Actual update 4500 of 5000, Stored update 4500 of 5000
Actual update 4550 of 5000, Stored update 4550 of 5000
Actual update 4600 of 5000, Stored update 4600 of 5000
Actual update 4650 of 5000, Stored update 4650 of 5000
Actual update 4700 of 5000, Stored update 4700 of 5000
Actual update 4750 of 5000, Stored update 4750 of 5000 --- Begin MLwiN error log ---
MLN - Software for N-level analysis. Mon 20 Oct 2025 08:54:00
/var/folders/1l/w0qjrymd3hb44s3skrzn4w4w0000gn/T//St94012.00000a
/var/folders/1l/w0qjrymd3hb44s3skrzn4w4w0000gn/T//St94012.00000e
/var/folders/1l/w0qjrymd3hb44s3skrzn4w4w0000gn/T//St94012.00000f
/var/folders/1l/w0qjrymd3hb44s3skrzn4w4w0000gn/T//St94012.000008
--- End MLwiN error log ---
Actual update 4800 of 5000, Stored update 4800 of 5000
Actual update 4850 of 5000, Stored update 4850 of 5000
Actual update 4900 of 5000, Stored update 4900 of 5000
Actual update 4950 of 5000, Stored update 4950 of 5000
Actual update 5000 of 5000, Stored update 5000 of 5000
Execution completedWarning: getversion plugin could not be loaded
MLwiN 3.16 multilevel model Number of obs = 1905
Multivariate response model (hierarchical)
Estimation algorithm: MCMC
-----------------------------------------------------------
| No. of Observations per Group
Level Variable | Groups Minimum Average Maximum
----------------+------------------------------------------
school | 73 2 26.1 104
-----------------------------------------------------------
Burnin = 500
Chain = 5000
Thinning = 1
Run time (seconds) = 1.11
Deviance (dbar) = 29388.29
Deviance (thetabar) = 28877.57
Effective no. of pars (pd) = 510.73
Bayesian DIC = 29899.02
------------------------------------------------------------------------------
| Mean Std. Dev. ESS P [95% Cred. Interval]
-------------+----------------------------------------------------------------
written |
cons_1 | 49.50607 .9899419 242 0.000 47.57032 51.46915
female_1 | -2.492796 .5577673 3451 0.000 -3.593134 -1.390972
-------------+----------------------------------------------------------------
csework |
cons_2 | 69.6792 1.232286 222 0.000 67.27528 72.03018
female_2 | 6.761519 .67411 3672 0.000 5.458935 8.089689
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects Parameters | Mean Std. Dev. ESS [95% Cred. Int]
-----------------------------+------------------------------------------------
Level 2: school |
var(cons_1) | 50.08992 10.5095 2376 32.88132 73.71481
cov(cons_1,cons_2) | 26.20564 9.578987 2490 9.893032 46.83127
var(cons_2) | 78.90363 15.49003 2637 53.52962 112.7813
-----------------------------+------------------------------------------------
Level 1: student |
var(cons_1) | 124.8802 4.356383 3734 116.8696 133.6906
cov(cons_1,cons_2) | 73.189 4.186575 3819 65.23343 81.54988
var(cons_2) | 180.573 6.35916 4068 168.6235 193.2279
------------------------------------------------------------------------------
------------------------------------------------------------------------------
| Mean Std. Dev. ESS P [95% Cred. Interval]
-------------+----------------------------------------------------------------
beta_cons_2 | .3318915 .1002958 2477 0.000 .1324359 .5296823
------------------------------------------------------------------------------Example 3: based on xtmixed helpfile
webuse nlswork, clear
version 12: xtmixed ln_w grade age c.age#c.age ttl_exp tenure ///
c.tenure#c.tenure || idcode: tenure, cov(uns) var
version 12: reffadjust4nlcom _cons tenure, eqn(idcode)
nlcom `r(beta_tenure)'(National Longitudinal Survey of Young Women, 14-24 years old in 1968)
Performing EM optimization:
Performing gradient-based optimization:
Iteration 0: Log likelihood = -8985.3816
Iteration 1: Log likelihood = -8966.2878
Iteration 2: Log likelihood = -8965.819
Iteration 3: Log likelihood = -8965.819
Computing standard errors:
Mixed-effects ML regression Number of obs = 28,099
Group variable: idcode Number of groups = 4,697
Obs per group:
min = 1
avg = 6.0
max = 15
Wald chi2(6) = 6767.13
Log likelihood = -8965.819 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
ln_wage | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
grade | .0690318 .0017885 38.60 0.000 .0655264 .0725372
age | .0321872 .0027908 11.53 0.000 .0267174 .037657
|
c.age#c.age | -.0006574 .0000466 -14.09 0.000 -.0007488 -.000566
|
ttl_exp | .0350762 .0011352 30.90 0.000 .0328513 .037301
tenure | .0393576 .0017198 22.88 0.000 .0359868 .0427284
|
c.tenure#|
c.tenure | -.0019926 .0001232 -16.18 0.000 -.0022341 -.0017512
|
_cons | .162264 .044888 3.61 0.000 .0742851 .2502429
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects parameters | Estimate Std. err. [95% conf. interval]
-----------------------------+------------------------------------------------
idcode: Unstructured |
var(tenure) | .0005174 .0000566 .0004176 .0006411
var(_cons) | .0632497 .0020808 .0593001 .0674624
cov(tenure,_cons) | .0007165 .0002809 .0001658 .0012671
-----------------------------+------------------------------------------------
var(Residual) | .0815016 .0007978 .0799529 .0830803
------------------------------------------------------------------------------
LR test vs. linear model: chi2(3) = 9168.36 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
_nl_1: tanh([atr1_1_1_2]_cons)*exp([lns1_1_1]_cons + [lns1_1_2]_cons)/ex
> p(2*[lns1_1_1]_cons)
------------------------------------------------------------------------------
ln_wage | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
_nl_1 | 1.384766 .6149302 2.25 0.024 .179525 2.590007
------------------------------------------------------------------------------mixed ln_w grade age c.age#c.age ttl_exp tenure c.tenure#c.tenure || ///
idcode: tenure, cov(uns)
reffadjust4nlcom _cons tenure, eqn(idcode)
nlcom `r(beta_tenure)'
Performing EM optimization ...
Performing gradient-based optimization:
Iteration 0: Log likelihood = -8985.3816
Iteration 1: Log likelihood = -8966.3961
Iteration 2: Log likelihood = -8965.819
Iteration 3: Log likelihood = -8965.819
Computing standard errors ...
Mixed-effects ML regression Number of obs = 28,099
Group variable: idcode Number of groups = 4,697
Obs per group:
min = 1
avg = 6.0
max = 15
Wald chi2(6) = 6767.13
Log likelihood = -8965.819 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
ln_wage | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
grade | .0690318 .0017885 38.60 0.000 .0655264 .0725372
age | .0321872 .0027908 11.53 0.000 .0267174 .037657
|
c.age#c.age | -.0006574 .0000466 -14.09 0.000 -.0007488 -.000566
|
ttl_exp | .0350762 .0011352 30.90 0.000 .0328513 .037301
tenure | .0393576 .0017198 22.88 0.000 .0359868 .0427284
|
c.tenure#|
c.tenure | -.0019926 .0001232 -16.18 0.000 -.0022341 -.0017512
|
_cons | .162264 .044888 3.61 0.000 .0742851 .2502429
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects parameters | Estimate Std. err. [95% conf. interval]
-----------------------------+------------------------------------------------
idcode: Unstructured |
var(tenure) | .0005174 .0000566 .0004176 .0006411
var(_cons) | .0632497 .0020808 .0593001 .0674624
cov(tenure,_cons) | .0007165 .0002809 .0001658 .0012671
-----------------------------+------------------------------------------------
var(Residual) | .0815016 .0007978 .0799529 .0830803
------------------------------------------------------------------------------
LR test vs. linear model: chi2(3) = 9168.36 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
_nl_1: tanh([atr1_1_1_2]_cons)*exp([lns1_1_1]_cons + [lns1_1_2]_cons)/ex
> p(2*[lns1_1_1]_cons)
------------------------------------------------------------------------------
ln_wage | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
_nl_1 | 1.384767 .6149298 2.25 0.024 .1795264 2.590007
------------------------------------------------------------------------------Example 4: based on xtmelogit helpfile
webuse bangladesh, clear
version 12: xtmelogit c_use urban age child* || district: urban, cov(uns) var
version 12: reffadjust4nlcom _cons urban, eqn(district)
nlcom `r(beta_urban)'(Bangladesh Fertility Survey, 1989)
note: children omitted because of collinearity.
Refining starting values:
Iteration 0: Log likelihood = -1215.8594 (not concave)
Iteration 1: Log likelihood = -1204.0802
Iteration 2: Log likelihood = -1199.7968
Performing gradient-based optimization:
Iteration 0: Log likelihood = -1199.7968
Iteration 1: Log likelihood = -1199.4726
Iteration 2: Log likelihood = -1199.3158
Iteration 3: Log likelihood = -1199.315
Iteration 4: Log likelihood = -1199.315
Mixed-effects logistic regression Number of obs = 1,934
Group variable: district Number of groups = 60
Obs per group:
min = 2
avg = 32.2
max = 118
Integration points = 7 Wald chi2(5) = 97.50
Log likelihood = -1199.315 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
c_use | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
urban | .8157872 .1715519 4.76 0.000 .4795516 1.152023
age | -.026415 .008023 -3.29 0.001 -.0421398 -.0106902
child1 | 1.13252 .1603285 7.06 0.000 .818282 1.446758
child2 | 1.357739 .1770522 7.67 0.000 1.010724 1.704755
child3 | 1.353827 .1828801 7.40 0.000 .9953882 1.712265
children | 0 (omitted)
_cons | -1.71165 .1605617 -10.66 0.000 -2.026345 -1.396954
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects parameters | Estimate Std. err. [95% conf. interval]
-----------------------------+------------------------------------------------
district: Unstructured |
var(urban) | .6663221 .3224714 .2580709 1.7204
var(_cons) | .3897434 .1292458 .2034723 .7465387
cov(urban,_cons) | -.4058846 .1755418 -.7499402 -.0618289
------------------------------------------------------------------------------
LR test vs. logistic model: chi2(3) = 58.42 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
_nl_1: tanh([atr1_1_1_2]_cons)*exp([lns1_1_1]_cons + [lns1_1_2]_cons)/ex
> p(2*[lns1_1_1]_cons)
------------------------------------------------------------------------------
c_use | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
_nl_1 | -.6091417 .1605193 -3.79 0.000 -.9237537 -.2945297
------------------------------------------------------------------------------meqrlogit c_use urban age child* || district: urban, cov(uns)
reffadjust4nlcom _cons urban, eqn(district)
nlcom `r(beta_urban)'note: children omitted because of collinearity.
Refining starting values:
Iteration 0: Log likelihood = -1215.8594 (not concave)
Iteration 1: Log likelihood = -1204.0802
Iteration 2: Log likelihood = -1199.7968
Performing gradient-based optimization:
Iteration 0: Log likelihood = -1199.7968
Iteration 1: Log likelihood = -1199.4726
Iteration 2: Log likelihood = -1199.3158
Iteration 3: Log likelihood = -1199.315
Iteration 4: Log likelihood = -1199.315
Mixed-effects logistic regression Number of obs = 1,934
Group variable: district Number of groups = 60
Obs per group:
min = 2
avg = 32.2
max = 118
Integration points = 7 Wald chi2(5) = 97.50
Log likelihood = -1199.315 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
c_use | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
urban | .8157872 .1715519 4.76 0.000 .4795516 1.152023
age | -.026415 .008023 -3.29 0.001 -.0421398 -.0106902
child1 | 1.13252 .1603285 7.06 0.000 .818282 1.446758
child2 | 1.357739 .1770522 7.67 0.000 1.010724 1.704755
child3 | 1.353827 .1828801 7.40 0.000 .9953882 1.712265
children | 0 (omitted)
_cons | -1.71165 .1605617 -10.66 0.000 -2.026345 -1.396954
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects parameters | Estimate Std. err. [95% conf. interval]
-----------------------------+------------------------------------------------
district: Unstructured |
var(urban) | .6663221 .3224714 .2580709 1.7204
var(_cons) | .3897434 .1292458 .2034723 .7465387
cov(urban,_cons) | -.4058846 .1755418 -.7499402 -.0618289
------------------------------------------------------------------------------
LR test vs. logistic model: chi2(3) = 58.42 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
_nl_1: tanh([atr1_1_1_2]_cons)*exp([lns1_1_1]_cons + [lns1_1_2]_cons)/ex
> p(2*[lns1_1_1]_cons)
------------------------------------------------------------------------------
c_use | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
_nl_1 | -.6091417 .1605193 -3.79 0.000 -.9237537 -.2945297
------------------------------------------------------------------------------Example 5: based on xtmepoisson helpfile
webuse epilepsy, clear
version 12: xtmepoisson seizures treat lbas lbas_trt lage visit || ///
subject: visit, cov(uns) var intpoints(9)
version 12: reffadjust4nlcom _cons visit, eqn(subject)
nlcom `r(beta_visit)'(Epilepsy data; progabide drug treatment)
Refining starting values:
Iteration 0: Log likelihood = -672.17188 (not concave)
Iteration 1: Log likelihood = -660.46056
Iteration 2: Log likelihood = -655.86888
Performing gradient-based optimization:
Iteration 0: Log likelihood = -655.86888
Iteration 1: Log likelihood = -655.68217
Iteration 2: Log likelihood = -655.68103
Iteration 3: Log likelihood = -655.68103
Mixed-effects Poisson regression Number of obs = 236
Group variable: subject Number of groups = 59
Obs per group:
min = 4
avg = 4.0
max = 4
Integration points = 9 Wald chi2(5) = 115.56
Log likelihood = -655.68103 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
seizures | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
treat | -.9286588 .4021638 -2.31 0.021 -1.716885 -.1404323
lbas | .8849767 .1312519 6.74 0.000 .6277277 1.142226
lbas_trt | .3379757 .2044442 1.65 0.098 -.0627276 .738679
lage | .4767192 .353622 1.35 0.178 -.2163672 1.169806
visit | -.2664098 .1647096 -1.62 0.106 -.5892347 .0564151
_cons | 2.099555 .2203709 9.53 0.000 1.667636 2.531474
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects parameters | Estimate Std. err. [95% conf. interval]
-----------------------------+------------------------------------------------
subject: Unstructured |
var(visit) | .5314808 .2293851 .2280931 1.238406
var(_cons) | .2514928 .0587892 .1590552 .3976522
cov(visit,_cons) | .0028715 .0887018 -.1709808 .1767238
------------------------------------------------------------------------------
LR test vs. Poisson model: chi2(3) = 324.54 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
_nl_1: tanh([atr1_1_1_2]_cons)*exp([lns1_1_1]_cons + [lns1_1_2]_cons)/ex
> p(2*[lns1_1_1]_cons)
------------------------------------------------------------------------------
seizures | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
_nl_1 | .0054029 .1669653 0.03 0.974 -.3218432 .3326489
------------------------------------------------------------------------------meqrpoisson seizures treat lbas lbas_trt lage visit || subject: visit, ///
cov(uns) intpoints(9)
reffadjust4nlcom _cons visit, eqn(subject)
nlcom `r(beta_visit)'
Refining starting values:
Iteration 0: Log likelihood = -672.17188 (not concave)
Iteration 1: Log likelihood = -660.46056
Iteration 2: Log likelihood = -655.86888
Performing gradient-based optimization:
Iteration 0: Log likelihood = -655.86888
Iteration 1: Log likelihood = -655.68217
Iteration 2: Log likelihood = -655.68103
Iteration 3: Log likelihood = -655.68103
Mixed-effects Poisson regression Number of obs = 236
Group variable: subject Number of groups = 59
Obs per group:
min = 4
avg = 4.0
max = 4
Integration points = 9 Wald chi2(5) = 115.56
Log likelihood = -655.68103 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
seizures | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
treat | -.9286588 .4021638 -2.31 0.021 -1.716885 -.1404323
lbas | .8849767 .1312519 6.74 0.000 .6277277 1.142226
lbas_trt | .3379757 .2044442 1.65 0.098 -.0627276 .738679
lage | .4767192 .353622 1.35 0.178 -.2163672 1.169806
visit | -.2664098 .1647096 -1.62 0.106 -.5892347 .0564151
_cons | 2.099555 .2203709 9.53 0.000 1.667636 2.531474
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects parameters | Estimate Std. err. [95% conf. interval]
-----------------------------+------------------------------------------------
subject: Unstructured |
var(visit) | .5314808 .2293851 .2280931 1.238406
var(_cons) | .2514928 .0587892 .1590552 .3976522
cov(visit,_cons) | .0028715 .0887018 -.1709808 .1767238
------------------------------------------------------------------------------
LR test vs. Poisson model: chi2(3) = 324.54 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
_nl_1: tanh([atr1_1_1_2]_cons)*exp([lns1_1_1]_cons + [lns1_1_2]_cons)/ex
> p(2*[lns1_1_1]_cons)
------------------------------------------------------------------------------
seizures | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
_nl_1 | .0054029 .1669653 0.03 0.974 -.3218432 .3326489
------------------------------------------------------------------------------Example 6: repeated group variable
webuse nlswork, clear
version 12: xtmixed ln_w grade age || idcode: tenure union, cov(uns) || ///
idcode: race, cov(uns) var
version 12: reffadjust4nlcom tenure union, eqn(idcode) sub(1)
nlcom `r(beta_union)'
version 12: reffadjust4nlcom race _cons, eqn(idcode) sub(2)
nlcom `r(beta__cons)'(National Longitudinal Survey of Young Women, 14-24 years old in 1968)
Performing EM optimization:
Performing gradient-based optimization:
Iteration 0: Log likelihood = -5208.9293
Iteration 1: Log likelihood = -5196.8989 (not concave)
Iteration 2: Log likelihood = -5176.1013
Iteration 3: Log likelihood = -5174.4196
Iteration 4: Log likelihood = -5171.2001
Iteration 5: Log likelihood = -5171.0595
Iteration 6: Log likelihood = -5171.0493
Iteration 7: Log likelihood = -5171.0492
Computing standard errors:
Mixed-effects ML regression Number of obs = 19,008
Group variable: idcode Number of groups = 4,132
Obs per group:
min = 1
avg = 4.6
max = 12
Wald chi2(2) = 2164.18
Log likelihood = -5171.0492 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
ln_wage | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
grade | .083625 .0022067 37.90 0.000 .0793 .08795
age | .0104518 .0004215 24.80 0.000 .0096256 .0112779
_cons | .2739857 .0306868 8.93 0.000 .2138407 .3341308
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects parameters | Estimate Std. err. [95% conf. interval]
-----------------------------+------------------------------------------------
idcode: Unstructured |
var(tenure) | .0007758 .0000666 .0006556 .0009181
var(union) | .0514475 .0041886 .0438594 .0603483
cov(tenure,union) | .0016667 .0004909 .0007044 .0026289
-----------------------------+------------------------------------------------
idcode: Unstructured |
var(race) | .0202217 .0139137 .0052498 .0778921
var(_cons) | .1578058 .031278 .1070066 .2327211
cov(race,_cons) | -.0401861 .022046 -.0833954 .0030232
-----------------------------+------------------------------------------------
var(Residual) | .0583641 .0007602 .0568929 .0598732
------------------------------------------------------------------------------
LR test vs. linear model: chi2(6) = 10154.21 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
_nl_1: tanh([atr1_1_1_2]_cons)*exp([lns1_1_1]_cons + [lns1_1_2]_cons)/ex
> p(2*[lns1_1_2]_cons)
------------------------------------------------------------------------------
ln_wage | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
_nl_1 | .0323954 .0101649 3.19 0.001 .0124725 .0523182
------------------------------------------------------------------------------
_nl_1: tanh([atr1_2_1_2]_cons)*exp([lns1_2_1]_cons + [lns1_2_2]_cons)/ex
> p(2*[lns1_2_2]_cons)
------------------------------------------------------------------------------
ln_wage | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
_nl_1 | -.2546555 .090066 -2.83 0.005 -.4311817 -.0781294
------------------------------------------------------------------------------mixed ln_w grade age || idcode: tenure union, cov(uns) || idcode: race, ///
cov(uns)
reffadjust4nlcom tenure union, eqn(idcode) sub(1)
nlcom `r(beta_union)'
reffadjust4nlcom race _cons, eqn(idcode) sub(2)
nlcom `r(beta__cons)'
Performing EM optimization ...
Performing gradient-based optimization:
Iteration 0: Log likelihood = -5208.9293
Iteration 1: Log likelihood = -5199.2605 (not concave)
Iteration 2: Log likelihood = -5175.53 (not concave)
Iteration 3: Log likelihood = -5173.8245
Iteration 4: Log likelihood = -5171.2707
Iteration 5: Log likelihood = -5171.0525
Iteration 6: Log likelihood = -5171.0492
Iteration 7: Log likelihood = -5171.0492
Computing standard errors ...
Mixed-effects ML regression Number of obs = 19,008
Group variable: idcode Number of groups = 4,132
Obs per group:
min = 1
avg = 4.6
max = 12
Wald chi2(2) = 2164.18
Log likelihood = -5171.0492 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
ln_wage | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
grade | .083625 .0022067 37.90 0.000 .0793 .08795
age | .0104518 .0004215 24.80 0.000 .0096256 .0112779
_cons | .2739857 .0306868 8.93 0.000 .2138407 .3341307
------------------------------------------------------------------------------
------------------------------------------------------------------------------
Random-effects parameters | Estimate Std. err. [95% conf. interval]
-----------------------------+------------------------------------------------
idcode: Unstructured |
var(tenure) | .0007758 .0000666 .0006556 .0009181
var(union) | .0514475 .0041886 .0438595 .0603483
cov(tenure,union) | .0016667 .0004909 .0007044 .0026289
-----------------------------+------------------------------------------------
idcode: Unstructured |
var(race) | .0202217 .0139134 .0052499 .07789
var(_cons) | .1578059 .0312773 .1070075 .2327191
cov(race,_cons) | -.0401862 .0220455 -.0833945 .0030222
-----------------------------+------------------------------------------------
var(Residual) | .0583641 .0007602 .0568929 .0598732
------------------------------------------------------------------------------
LR test vs. linear model: chi2(6) = 10154.21 Prob > chi2 = 0.0000
Note: LR test is conservative and provided only for reference.
_nl_1: tanh([atr1_1_1_2]_cons)*exp([lns1_1_1]_cons + [lns1_1_2]_cons)/ex
> p(2*[lns1_1_2]_cons)
------------------------------------------------------------------------------
ln_wage | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
_nl_1 | .0323953 .0101649 3.19 0.001 .0124726 .0523181
------------------------------------------------------------------------------
_nl_1: tanh([atr1_2_1_2]_cons)*exp([lns1_2_1]_cons + [lns1_2_2]_cons)/ex
> p(2*[lns1_2_2]_cons)
------------------------------------------------------------------------------
ln_wage | Coefficient Std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
_nl_1 | -.2546557 .0900641 -2.83 0.005 -.431178 -.0781333
------------------------------------------------------------------------------