[R-sig-ME] Lmer- Convergence issues
arun
smartpink111 at yahoo.com
Wed May 9 19:21:04 CEST 2012
Dear R mixed group,
Last week I posted a convergence issue (Warning message:In mer_finalize(ans) : false convergence (8)) while I was running one of my models.
(fmNC<-lmer(Response~Wavelength*Start_Resp*time+(1|resid)+(1+time|Subject)+(1|RepNest), verbose=TRUE,family=binomial,data=BehavdatOrig2)).
Then I came across David Hugh's blog (http://davidhughjones.blogspot.com/2009/11/lme-false-convergence.html) about the lmer false convergence.
I divided the variable time by 10 (as it's estimate for beta was the smallest) and called it a new variable time1. When I run the new model,
(fm<-lmer(Response~Wavelength*Start_Resp*time1+(1|resid)+(1+time1|Subject)+(1|RepNest), verbose=TRUE,family=binomial,data=BehavdatOrig2)), the convergence issue was solved and I am getting estimates of coefficients closer to what I get in other models. Only difference in the coefficients were in the estimates and standard errors involving time. It is now one decimal to the left (estimates for time I was getting previously in other models were 0.08, 0.086 with same z value and p value. Same is true for the interactions).
I guess I need to convert to one decimal right for the time variable when I report.
I tried several ways to converge (from other posts), but none of them worked except the division,
Any comments will be appreciated.
Thanking you,
A.K.
Generalized linear mixed model fit by the Laplace approximation
Formula: Response ~ Wavelength * Start_Resp * time1 + (1 | resid) + (1 + time1 | Subject) + (1 | RepNest)
Data: BehavdatOrig2
AIC BIC logLik deviance
1325 1420 -645.7 1291
Random effects:
Groups Name Variance Std.Dev. Corr
resid (Intercept) 6.9806e-11 8.3550e-06
RepNest (Intercept) 9.0018e+00 3.0003e+00
Subject (Intercept) 7.0247e+00 2.6504e+00
time1 1.2760e+01 3.5722e+00 -0.833
Number of obs: 1960, groups: resid, 1960; RepNest, 98; Subject, 98
Fixed effects:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -4.0456 0.9875 -4.097 4.19e-05 ***
WavelengthGreen 0.5634 1.4388 0.392 0.6954
WavelengthRed -5.1001 3.2495 -1.570 0.1165
Start_RespL 10.9094 1.9002 5.741 9.41e-09 ***
time1 0.8834 0.8693 1.016 0.3095
WavelengthGreen:Start_RespL -2.7382 2.4821 -1.103 0.2700
WavelengthRed:Start_RespL 1.9070 3.9939 0.477 0.6330
WavelengthGreen:time1 -0.9994 1.2802 -0.781 0.4350
WavelengthRed:time1 1.0487 2.4436 0.429 0.6678
Start_RespL:time1 -3.8931 1.5605 -2.495 0.0126 *
WavelengthGreen:Start_RespL:time1 -0.1811 2.1123 -0.086 0.9317
WavelengthRed:Start_RespL:time1 3.5655 3.2699 1.090 0.2755
More information about the R-sig-mixed-models
mailing list