Troubleshooting with glmmTMB

2017-02-16

This vignette covers common problems that occur while using glmmTMB. The contents will expand with experience.

If your problem is not covered below, try updating to the latest version of glmmTMB on GitHub. The developers might have solved the problem in a newer version.

Warnings

Model convergence problem; non-positive-definite Hessian matrix

You may see the same warning as in the following example:

zinbm0 = glmmTMB(count~spp + (1|site), zi=~spp, Salamanders, family="nbinom2")
## Warning in glmmTMB(count ~ spp + (1 | site), zi = ~spp, Salamanders, family
## = "nbinom2"): Model convergence problem; non-positive-definite Hessian
## matrix. See vignette('troubleshooting')

This often occurs when a model is overparameterized (i.e. the data does not contain information to estimate the parameters). In the example above, the problem is that zero observations do not depend on species as in zi=~spp; rather, they depend on another covariate that is missing from the model minded. Plotting the data against potential covariates should help to avoid fitting unreasonable models that don’t converge.

In some cases, scaling predictor variables may help.

Models with non-positive definite Hessian matricies should be excluded from further consideration, in general.

Model convergence problem; Eigen value problems

m1 = glmmTMB(count~spp + mined + (1|site), zi=~spp + mined, Salamanders, family="genpois")
## Warning in glmmTMB(count ~ spp + mined + (1 | site), zi = ~spp + mined, :
## Model convergence problem; extreme or very small eigen values detected. See
## vignette('troubleshooting')

In this example, the fixed effect covariance matrix is NaN. It may have to do with the generalized Poisson (genpois) distribution, which is known to have convergence problems; luckily, the negative binomial (nbinom1 and nbinom2) and/or Conway-Maxwell Poisson (compois) are good alternatives.

Models with convergence problems should be excluded from further consideration, in general.

NA/NaN function evaluation

Warning in nlminb(start = par, objective = fn, gradient = gr) :
  NA/NaN function evaluation

This warning often occurs when the optimizer wanders into a region of parameter space that is invalid. It is not a problem as long as the optimizer has left that region of parameter space upon convergence, which is indicated by an absence of the model convergence warnings described above.

The following warnings can be treated in the same way as an NA/NaN function evaluation:

Cholmod warning 'matrix not positive definite'
Warning in f(par, order = order, ...) : value out of range in 'lgamma'

Errors

NA/NaN gradient evaluation

dat1 = expand.grid(y=-1:1, rep=1:10)
m1 = glmmTMB(y~1, dat1, family="nbinom2")
## Error in nlminb(start = par, objective = fn, gradient = gr): NA/NaN gradient evaluation
## Timing stopped at: 0.004 0 0.005

The error occurs here because the negative binomial distribution is not appropriate for data with negative values.

gradient length

Error in nlminb(start = par, objective = fn, gradient = gr) : 
  gradient function must return a numeric vector of length x
Error in optimHess(par.fixed, obj$fn, obj$gr): 
  gradient in optim evaluated to length x

Try rescaling predictor variables. Try a simpler model and build up.