[R-sig-ME] (no subject)
DESPINA MICHAILIDOU
de@m|ch@|||dou @end|ng |rom gm@||@com
Wed May 1 19:24:10 CEST 2019
Hi,
I run the following regression analysis with lme4 package and I get the
following message.
glmm_CA_intens<- glmer(Dizz_today ~ CA_intens + (1 | ID/SCAN_DATE/Side),
data=TAK_data, family=binomial(link = "logit"))
summary(glmm_CA_intens)
Generalized linear mixed model fit by maximum likelihood (Laplace
Approximation) ['glmerMod']
Family: binomial ( logit )
Formula: Dizz_today ~ CA_intens + (1 | ID/SCAN_DATE/Side)
Data: TAK_data
AIC BIC logLik deviance df.resid
41.7 58.4 -15.8 31.7 205
Scaled residuals:
Min 1Q Median 3Q Max
-0.003892 -0.000827 -0.000826 0.000000 0.054166
Random effects:
Groups Name Variance Std.Dev.
Side:(SCAN_DATE:ID) (Intercept) 0.0 0.00
SCAN_DATE:ID (Intercept) 3223.4 56.78
ID (Intercept) 199.4 14.12
Number of obs: 210, groups: Side:(SCAN_DATE:ID), 210; SCAN_DATE:ID, 105;
ID, 55
Fixed effects:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.419e+01 1.901e+01 -0.747 0.455
CA_intens -2.892e+03 1.103e+07 0.000 1.000
Correlation of Fixed Effects:
(Intr)
CA_intens 0.000
convergence code: 0
boundary (singular) fit: see ?isSingular
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
2: In vcov.merMod(object, correlation = correlation, sigm = sig) :
variance-covariance matrix computed from finite-difference Hessian is
not positive definite or contains NA values: falling back to var-cov
estimated from RX
>
What is wrong?
Thank you in advance for your help.
Sincerely,
Despina
[[alternative HTML version deleted]]
More information about the R-sig-mixed-models
mailing list