Displaying 1 result from an estimated 1 matches for "bhrt".
Did you mean:
bert
2010 Sep 24
1
Fitting GLMM models with glmer
...ation to Fit Nonlinear Mixed-Models
with Nonnormal Random Effects - 2006). Firstly the authors fit a
Poisson model with canonical link and a single normal random effect
bi ~ N(0;Sigma^2).The SAS routine was:
log_s =log(SURVT)
cens=1
proc nlmixed data=liver;
parms logsig2 = 0 b0 = -2.8 btrt = -0.54 bhrt =0.18;
xb= log_s + b0 + btrt * treat + bhrt * heart+ bi;
lambda=exp(xb);
model cens ~ poisson(lambda);
random bi ~ normal(0,exp(logsig2)) subject=INST;
run;
I obtained the same results for parameters estimates and
standarderrors, by using:
glmer(cens ~ treat + heart + offset(log(SURVT) + (1 | INS...