list.coef: Transform glm and mixed model objects into model summaries...

In catregs: Post-Estimation Functions for Generalized Linear Mixed Models

Transform glm and mixed model objects into model summaries that include coefficients, standard errors, exponentiated coefficients, confidence intervals and percent change.

Description

Given a glm model object or a mixed model model object, the function computes and returns: coefficients, standard errors, z-scores, confidence intervals, p-values, exponentiated coefficients, confidence intervals for exponentiated coefficients, and percent change.

Supported models include logistic regression via the glm function, ordinal regression via mass::polr, multinomial regression via nnet:multinom, Poisson regression via the glm function, negative binomial regression via mass::glm.nb, mixed effects models for continuous outcomes with serial correlation via nlme::lme, mixed effects logistic and poisson regression via lme4::glmer, mixed effects negative binomial regression via lme4::glmer.nb, and mixed effects ordinal regression via ordinal::clmm.

Usage

list.coef(model,rounded=3,alpha=.05)

Arguments

model	A model object. The model should be regression model for limited dependent variables, such as a logistic regression, or a mixed model from nlme or lme4/lmerTest.
rounded	The number of decimal places to round the output. The default is 3.
alpha	The alpha value for confidence intervals. Default is .05.

Value

b	The estimated model coefficients from the model object.
S.E.	The estimated model standard errors from the model object.
Z	The z-statistic corresponding to the coefficient.
LL CI	Given the coefficient, standard error and alpha value (default=.05), the lower limit of the confidence interval around the coefficient is reported.
UL CI	Given the coefficient, standard error and alpha value (default=.05), the upper limit of the confidence interval around the coefficient is reported.
p-val	The p-value associated with the z-statistic.
exp(b)	The exponentiated model coefficients. That is, odds ratios in the case of a logistic regression, or incidence rate ratios in the case of a count model.
LL CI for exp(b)	Given the exponentiated coefficient, standard error and alpha value (default=.05), the lower limit of the confidence interval around the exponentiated coefficient is reported.
UL CI for exp(b)	Given the exponentiated coefficient, standard error and alpha value (default=.05), the upper limit of the confidence interval around the exponentiated coefficient is reported.
Percent	The coefficients in terms of percent change. That is, 100*(exp(coef(model))-1)

Author(s)

David Melamed

Examples

data("Mize19AH")
m1 <- glm(alcB ~woman*parrole + age + race2 +
race3 + race4 + income + ed1 + ed2 + ed3 +
ed4,family="binomial",data=Mize19AH)
list.coef(m1,rounded=4)
list.coef(m1,rounded=4,alpha=.01)

catregs documentation built on Sept. 11, 2024, 6:18 p.m.