I am struggling with this specific mixed model which keeps failing to converge after trying different optimizers In the model the response variable is binary 01 with 4 numeric predictors and 3 random effects
In terms of which optimizer to use I would tend to prefer the fit with the best loglikelihood ie biggest for negative values this means closest to zero because for identical models but with different approximations here optimizers the best loglikelihood indicates the best approximation
r Model fails to converge in glmer after trying different
r Default lme4 optimizer requires lots of iterations for high
optimizer character name of optimizing functions A character vector or list of functions length 1 for lmer or glmer possibly length 2 for glmerBuiltin optimizers are NelderMead bobyqa from the minqa package nlminbwrap using base R s nlminb and the default for lmerControl nloptwrapAny minimizing function that allows box constraints can be used provided that it
glmer function RDocumentation
begingroup I dont have a full answer for you so Ill leave this as a comment In my experience glmer is quite slow especially for models that have a complex random effects structure eg many random slopes crossed random effects etc My first suggestion would be to try again with a simplified random effects structure However if youre experiencing this problem with a random
It says that we can and that we should compare results with results from other optimizers However you did not use a different optimizer bobyqa is the default for glmer but rather increased the maximum allowed number of function evaluations ie allowed more iterations for attempting to reach convergence
Glmer Optimizer
lme4 performance tips The Comprehensive R Archive Network
Mixed Effects Logistic Regression R Data Analysis Examples OARC Stats
r GLMM optimiser test optimxLBFGSB doesnt converge but the
We would like to show you a description here but the site wont allow us
Mixed Effects Modeling Tips Use a Fast Optimizer but Perform
All arguments have defaults and can be grouped into general control parameters most importantly optimizer further restartedge etc model or datachecking specifications in short checking options such as checknobsvsrankZ or checkrankX currently not for nlmerControl all the parameters to be passed to the optimizer eg
Glmer Optimizer
The optimizer does not influence the parameter estimates if these facets all have dots in a single column like this Here we can be assured that the models parameters we report do not depend on the optimizer we chose You could also do something similar with the zstatistics and tstatistics of a particular variable from the model Lets
lmerControl Control of Mixed Model Fitting R Package Documentation
Fit a generalized linear mixedeffects model GLMM Both fixed effects and random effects are specified via the model formula
r solution to the warning message using glmer Stack Overflow
lmerControl function RDocumentation
Below we use the glmer command to estimate a mixed effects logistic regression model with Il6 CRP and LengthofStay as patient level continuous predictors To avoid a warning of nonconvergence we specify a different optimizer with the argument controlglmerControloptimizerbobyqa Although the model will produce nearly identical