Hello,

I am trying to fit the following model:

stan_glmer(formula(y~logDose*sampleName+(1|block)+(0+logDose|block)),

family=binomial(link=“probit”),

data=dataSet, adapt_delta=0.99)

The simulated data is a logDose/Response dataset with 5 doses and the responses are 0’s and 1’s, with two samples and two blocks, and 20 replicates at each dose.

I am running into convergence issues;

Warning messages:

1: There were 43 divergent transitions after warmup. Increasing adapt_delta above 0.999 may help. See

http://mc-stan.org/misc/warnings.html#divergent-transitions-after-warmup

2: Examine the pairs() plot to diagnose sampling problems

The RHat and effective sample space metrics both give results that are consistent with convergence, whereas the divergent transitions do not. I looked at the pairs plot and the divergent transitions are all “below the diagonal”.

I am happy to provide the data.

I am just not clear on how effective reparameterization happens with an rstanarm model, or if you have any other suggestions.

Thank you in advance,

Ramiro

- Operating System: Centos 7
- rstanarm Version: 2.17.3