I have troubles computing a power analysis for a GLMM model. The code is:
`power <- mixedpower(model = glm_model, data= df,
fixed_effects=c("task","height","length","color"), # fixed effects
simvar = "subject", # random effects
steps = c(20,40,60), # sample sizes to simulate
critical_value = 2,# reflects alpha level of 5%
SESOI = effects, # smallest effect size of interest
n_sim = 100)`
I use Gamma distribution with a log link. The errors I get are:
`"Simulations for step 20are based on 0 successful single runs"
Error in apply(store_simulations, MARGIN = 1, FUN = mean, na.rm = T) :
dim(X) must have positive length
In addition: Warning messages:
1: In vcov.merMod(object, use.hessian = use.hessian) :
variance-covariance matrix computed from finite-difference Hessian is
not positive definite or contains NA values: falling back to var-cov estimated from RX`
However, if I use the same model as a linear model, it works. What could be the reason for that?