I'm using Stata 17 on Windows 10 and the Stata command xrigls to determine reference intervals. I should be grateful if someone could tell me how to get the equation of the model from the output (I found the explanation in the Stata Bulletin STB-38 somewhat confusing). For example the output of one of my analyses is below.
Many thanks
. xrigls AS_mean crl , fp (m:df 4, s: df 2) detail centile (5 50 95) cycles(3)
--- FP Powers ---
Cycle Mean SD Deviance Change Residual SS
-----------------------------------------------------------------
0 3 3 1641.665 0.000 2015.069
1 2 3 1 1590.287 -51.378 1445.612
2 2 3 1 1590.422 0.135 1442.72
3 2 3 1 1590.422 0.000 1442.719
Final deviance = 1590.422 (360 observations).
Power(s) for mean curve = 2 3. Power(s) for SD curve = 1.
Regression for mean curve
-------------------------
(sum of wgt is 108.1984317153692)
Source | SS df MS Number of obs = 360
-------------+---------------------------------- F(2, 357) = 1700.06
Model | 13740.6695 2 6870.33475 Prob > F = 0.0000
Residual | 1442.71866 357 4.04122874 R-squared = 0.9050
-------------+---------------------------------- Adj R-squared = 0.9044
Total | 15183.3882 359 42.2935604 Root MSE = 2.0103
------------------------------------------------------------------------------
AS_mean | Coefficient Std. err. t P>|t| [95% conf. interval]
-------------+----------------------------------------------------------------
Xm_1 | 6.367892 .3408574 18.68 0.000 5.697552 7.038233
Xm_2 | -1.06975 .1179238 -9.07 0.000 -1.301663 -.8378374
_cons | 3.210562 .337609 9.51 0.000 2.546609 3.874514
------------------------------------------------------------------------------
Variable Storage Display Value
name type format label Variable label
-------------------------------------------------------------------------------------------------------------
Xm_1 double %10.0g X^2: X = crl/10
Xm_2 double %10.0g X^3: X = crl/10
Regression for SD curve
-----------------------
Source | SS df MS Number of obs = 360
-------------+---------------------------------- F(1, 358) = 33.78
Model | 136.458984 1 136.458984 Prob > F = 0.0000
Residual | 1446.12032 358 4.03944224 R-squared = 0.0862
-------------+---------------------------------- Adj R-squared = 0.0837
Total | 1582.57931 359 4.4082989 Root MSE = 2.0098
------------------------------------------------------------------------------
AS_mean | Coefficient Std. err. t P>|t| [95% conf. interval]
-------------+----------------------------------------------------------------
crl | .1043137 .0179474 5.81 0.000 .0690182 .1396092
_cons | .189176 .3454523 0.55 0.584 -.4901947 .8685468
------------------------------------------------------------------------------
Variable Storage Display Value
name type format label Variable label
-------------------------------------------------------------------------------------------------------------
crl double %12.0g Crown-rump length (mm)
Many thanks
. xrigls AS_mean crl , fp (m:df 4, s: df 2) detail centile (5 50 95) cycles(3)
--- FP Powers ---
Cycle Mean SD Deviance Change Residual SS
-----------------------------------------------------------------
0 3 3 1641.665 0.000 2015.069
1 2 3 1 1590.287 -51.378 1445.612
2 2 3 1 1590.422 0.135 1442.72
3 2 3 1 1590.422 0.000 1442.719
Final deviance = 1590.422 (360 observations).
Power(s) for mean curve = 2 3. Power(s) for SD curve = 1.
Regression for mean curve
-------------------------
(sum of wgt is 108.1984317153692)
Source | SS df MS Number of obs = 360
-------------+---------------------------------- F(2, 357) = 1700.06
Model | 13740.6695 2 6870.33475 Prob > F = 0.0000
Residual | 1442.71866 357 4.04122874 R-squared = 0.9050
-------------+---------------------------------- Adj R-squared = 0.9044
Total | 15183.3882 359 42.2935604 Root MSE = 2.0103
------------------------------------------------------------------------------
AS_mean | Coefficient Std. err. t P>|t| [95% conf. interval]
-------------+----------------------------------------------------------------
Xm_1 | 6.367892 .3408574 18.68 0.000 5.697552 7.038233
Xm_2 | -1.06975 .1179238 -9.07 0.000 -1.301663 -.8378374
_cons | 3.210562 .337609 9.51 0.000 2.546609 3.874514
------------------------------------------------------------------------------
Variable Storage Display Value
name type format label Variable label
-------------------------------------------------------------------------------------------------------------
Xm_1 double %10.0g X^2: X = crl/10
Xm_2 double %10.0g X^3: X = crl/10
Regression for SD curve
-----------------------
Source | SS df MS Number of obs = 360
-------------+---------------------------------- F(1, 358) = 33.78
Model | 136.458984 1 136.458984 Prob > F = 0.0000
Residual | 1446.12032 358 4.03944224 R-squared = 0.0862
-------------+---------------------------------- Adj R-squared = 0.0837
Total | 1582.57931 359 4.4082989 Root MSE = 2.0098
------------------------------------------------------------------------------
AS_mean | Coefficient Std. err. t P>|t| [95% conf. interval]
-------------+----------------------------------------------------------------
crl | .1043137 .0179474 5.81 0.000 .0690182 .1396092
_cons | .189176 .3454523 0.55 0.584 -.4901947 .8685468
------------------------------------------------------------------------------
Variable Storage Display Value
name type format label Variable label
-------------------------------------------------------------------------------------------------------------
crl double %12.0g Crown-rump length (mm)
