[R-sig-ME] lmer Control Parameters Question

David Duffy David.Duffy at qimr.edu.au
Mon Jul 21 23:07:10 CEST 2014


On Tue, 22 Jul 2014, Douglas Bates wrote:

>> I am working with the latest version of lme4 and having a great deal of
>> difficulty specify control parameters (specifically, iterations and
>> convergence criteria).  I have been using variants of the following, with
>> no success:

The usable arguments in optCtrl vary according to which optimizer you have 
chosen:

glmer(cbind(incidence, size - incidence) ~ period + (1 | herd),
   family = binomial, data = cbpp,
   control=glmerControl(optimizer="Nelder_Mead",
                        optCtrl=list(FtolAbs = 1e-2, iprint=1),
                        boundary.tol = 1e-2))
load(optimx)
glmer(cbind(incidence, size - incidence) ~ period + (1 | herd),
   family = binomial, data = cbpp,
   control=glmerControl(optimizer="optimx",
                        optCtrl=list(method="nlminb", trace=1, kkt=FALSE),
                        boundary.tol = 1e-2))



| David Duffy (MBBS PhD)
| email: David.Duffy at qimrberghofer.edu.au  ph: INT+61+7+3362-0217 fax: -0101
| Genetic Epidemiology, QIMR Berghofer Institute of Medical Research
| 300 Herston Rd, Brisbane, Queensland 4006, Australia  GPG 4D0B994A



More information about the R-sig-mixed-models mailing list