Closed jeroenclaes closed 9 years ago
Would it be possible to send me some sample data, or the fitted model object saved from the environment (save(model_LH, file = "model_lh.RData")
), via email? I haven't worked with multiple random intercepts and slopes yet, so there might be a bug when plotting predicted probabilities for various random intercepts...
Which version of the package sjPlot do you use? I can't reproduce this bug, all plots are shown (however, a takes very long due to many coefficients and grouping levels).
I tried both the cran version and the developer github version, both gave me the same issue with the code on the package website. I will try on Monday to update the package and I will let you know if that solves the issue.
Thanks for looking into it.
Jeroen Claes, PhD Bresstraat 14/2 2018 Antwerpen
Sent from a mobile device. Please excuse errors and/or brevity Op 28-mrt.-2015 18:47 schreef "Daniel Lüdecke" notifications@github.com:
Which version of the package sjPlot do you use? I can't reproduce this bug, all plots are shown (however, a takes very long due to many coefficients and grouping levels).
Reply to this email directly or view it on GitHub https://github.com/sjPlot/devel/issues/13#issuecomment-87273256.
What about the other packages? Are lme4 and arm up to date?
I updated the lme4 and arm package and re-ran the code. I get the same error:
Error in $<-.data.frame
(*tmp*
, "xbeta", value = numeric(0)) :
replacement has 0 rows, data has 2093
which version of sjPlot do you exactly use? Do you get a traceback / error call with some information at which point (code line) in the function call the error occurs? RStudio sometimes offers you a traceback on error calls.
I updated to 1.6 and updated all other packages too, it seems to work now. Thanks for looking into it!
Ok, so I may close this issue? Btw, latest sjPlot should be 1.7 (perhaps it was a typo?).
I also recommend using the vars
parameter and plot selected variables only at first. Plotting all coefficients for all random intercept levels may take some time. ;-)
I have a model with the following formula:
model_LH<- glmer(Tipo ~ Co.2.Pr + Education + Negation + Pr.2.Pr + Typical.Action.Chain.Pos + Style*Tense +(1 + Co.2.Pr | Muestra)+ (1 + Negation | Muestra) +(1 + Typical.Action.Chain.Pos| Muestra)+ (1 | Nombrelema), data=datalh, family="binomial", control=glmerControl(optimizer="bobyqa"))
When I run sjp.glmer(model_LH, type = "ri.pc", show.se = TRUE)
I get the following error: Error in
$<-.data.frame
(*tmp*
, "xbeta", value = numeric(0)) : replacement has 0 rows, data has 2093