Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | Sofia Ramiro <sofiaramiro@hotmail.com> |
| To | Statalist <statalist@hsphsun2.harvard.edu> |
| Subject | st: Age-specific reference intervals (xrigls) - no equation for SD? |
| Date | Fri, 1 Jun 2012 15:54:19 +0100 |
Dear Statalist members, I am trying to define age-specific reference intervals with the command xrigls. However, when running it, for some of the variables for which I want to define the age intervals I get an equation for the mean (with the corresponding power), but no equation for the SD - example below. I don't understand why. Can any of you explain this? Are the values derived without the SD equation also right? I don't understand why in the 1st table (FP powers) the change appears always as 0. I inspected the data and there is variation in this variable and throughout the different ages. Thanks Sofia . xrigls schober15 age if sample==1, fp(m:df 4,s:df 2) centile(2.5 10 25 75 90 97.5) detail --- FP Powers --- Cycle Mean SD Deviance Change Residual SS ----------------------------------------------------------------- 0 1 1217.262 0.000 509.4298 1 1 1217.262 0.000 509.4298 2 1 1217.262 0.000 509.4298 Final deviance = 1217.262 (393 observations). Power(s) for mean curve = 1. Regression for mean curve ------------------------- (sum of wgt is 3.4287e+02) Source | SS df MS Number of obs = 393 -------------+------------------------------ F( 1, 391) = 24.05 Model | 31.3308787 1 31.3308787 Prob > F = 0.0000 Residual | 509.429806 391 1.30288953 R-squared = 0.0579 -------------+------------------------------ Adj R-squared = 0.0555 Total | 540.760685 392 1.37949154 Root MSE = 1.1414 ------------------------------------------------------------------------------ schober15 | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- age | -.0203814 .0041562 -4.90 0.000 -.0285528 -.01221 _cons | 7.191451 .1913316 37.59 0.000 6.815284 7.567619 ------------------------------------------------------------------------------ * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/