> library(foreign)
> ds <- read.dta("c:/ecg.dta")
> attach(ds)
> library(geepack)
> summary(geeglm(y ~ trt + period, data=ds, id = id, waves=period,
+ family=binomial("logit"), corstr="exch", std.err="san.se"))
Call:
geeglm(formula = y ~ trt + period, family = binomial("logit"),
data = ds, id = id, waves = period, corstr = "exch", std.err = "san.se")
Coefficients:
Estimate Std.err Wald Pr(>|W|)
(Intercept) -1.2348 0.2978 17.195 3.37e-05 ***
trt 0.5689 0.2365 5.788 0.0161 *
period 0.2950 0.2352 1.574 0.2096
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Estimated Scale Parameters:
Estimate Std.err
(Intercept) 0.9975 0.09558
Correlation: Structure = exchangeable Link = identity
Estimated Correlation Parameters:
Estimate Std.err
alpha 0.6243 0.1246
Number of clusters: 67 Maximum cluster size: 2
Mixed Effects Logistic Regression Model (Random Intercept)
> library(lme4)
> summary(glmer(y ~ trt + period + (1 | id), family=binomial, nAGQ=75))
Generalized linear mixed model fit by the adaptive Gaussian Hermite approximation
Formula: y ~ trt + period + (1 | id)
AIC BIC logLik deviance
177 188 -84.4 169
Random effects:
Groups Name Variance Std.Dev.
id (Intercept) 15.6 3.95
Number of obs: 134, groups: id, 67
Fixed effects:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -3.820 0.848 -4.50 6.7e-06 ***
trt 1.503 0.654 2.30 0.022 *
period 0.803 0.653 1.23 0.219
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) trt
trt -0.548
period -0.529 0.203