Stata 15 help for estat icc

[ME] estat icc -- Estimate intraclass correlations

Syntax

estat icc [, level(#)]

Menu for estat

Statistics > Postestimation

Description

estat icc is for use after estimation with mixed, meintreg, metobit, melogit, meqrlogit, and meprobit. estat icc is also for use after estimation with meglm in cases when the fitted model is a linear, logit, or probit mixed-effects model.

estat icc displays the intraclass correlation for pairs of responses at each nested level of the model. Intraclass correlations are available for random-intercept models or for random-coefficient models conditional on random-effects covariates being equal to 0. They are not available for crossed-effects models or with residual error structures other than independent structures.

Option

level(#) specifies the confidence level, as a percentage, for confidence intervals. The default is level(95) or as set by set level.

Example

Setup . webuse productivity . mixed gsp private emp hwy water other unemp || region: || state:

Compute residual intraclass correlations . estat icc

Stored results

estat icc stores the following in r():

Scalars r(icc#) level-# intraclass correlation r(se#) standard errors of level-# intraclass correlation r(level) confidence level of confidence intervals

Macros r(label#) label for level #

Matrices r(ci#) vector of confidence intervals (lower and upper) for level-# intraclass correlation

For a G-level nested model, # can be any integer between 2 and G.


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index