Hello,
I am using Stata-MP 13.1. I am using -melogit- to analyze two-level models with cluster robust standard errors, following which I would like to compute the intraclass correlation coefficient.
However, when I enter the -estat icc- command, I get the following error
When I use -xtlogit- instead of -melogit-, the output does provide the rho value.
Would you'll please help me understand why I should not be calculating an icc/rho after -melogit-
Thank you,
Caroline
I am using Stata-MP 13.1. I am using -melogit- to analyze two-level models with cluster robust standard errors, following which I would like to compute the intraclass correlation coefficient.
Code:
webuse bangladesh melogit c_use urban age child* || district:, or vce(cl district) estat icc
Code:
estat icc not allowed after estimation with robust standard errors r(321)".
Code:
webuse bangladesh xtset district xtlogit c_use urban age child*, or vce(cl district)
Thank you,
Caroline