Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Pagel, Christina" <c.pagel@ucl.ac.uk> |
To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
Subject | st: using the output from quadchk |
Date | Tue, 28 Aug 2012 17:04:48 +0000 |
I am using quadchk repeatedly to check the fits of a series of xtlogit comm= ands within a foreach loop in a DO file. What I would like to do is to writ= e the quadchk output to a matrix and then to check within the DO file wheth= er the relative differences are too large, outputting a warning for the use= r. This would save people having to check the quadchk output manually. Currently I save the output, say, xtlogit by a command like this: Matrix TempMatrix =3D r(table) And then write the relevant information to a matrix of my own, e.g.: Matrix Results[3,1]=3Dexp(TempMatrix[1,1])/(1+exp(TempMatrix[1,1])) And finally at the end of the DO file I show this matrix to the user: Matrix list Results. Doing ereturn list after running quadchk, I can't work out where the relati= ve differences are stored and where I should look for them! Ideally I'd lik= e to have code that went something like this: Quietly quadchk Matrix TempDiffs =3D XXXXXXXX If TempDiffs[X,y]>10^(-3) { Display "Warning: quadchk indicates fit unreliable" } * * 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/