Skip to content

Problems using lmer {lme4}

5 messages · Andrea Goijman, Bert Gunter, Patrick Coulombe

Bert Gunter
# 
It would be better to post this on the r-sig-mixed-models list, I think.

-- Bert

On Fri, May 17, 2013 at 10:02 AM, Andrea Goijman
<agoijman at cnia.inta.gov.ar> wrote:

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Andrea Goijman
    

    

      
      #
    

  


  

      
Thanks! I didn't know about that list. I forwarded my question  there

Andrea
Bert Gunter wrote

            

      
      
      
    

        

      
      
    

        

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

        

            

      
      
      
    

        

      
      
    

        
--
View this message in context: http://r.789695.n4.nabble.com/Problems-using-lmer-lme4-tp4667345p4667349.html
Sent from the R help mailing list archive at Nabble.com.

  
  


  


      

    

    

      
      

        


  

        

      Patrick Coulombe
    

    

      
      #
    

  


  

      
Hi Andrea,

I'm not exactly sure what you're trying to do, but you've included a
random effect for a "site" coefficient that's not even in your list of
fixed effects... you're basically allowing the coefficient for site to
vary across routes, but you're never including the coefficient in the
first place. Given the appropriate nested structure (I haven't looked
at your data), a model that should run would be:

m2 <- lmer(cbind(punto6,5) ~ sp + site + (site|route) ,family=binomial,data = d)

The problem here is I'm not sure this is what you mean to do... If you
could way simplify your example (for example, by including a small
number of mock observations with only the relevant variables, in a
table instead of in R syntax), that would be helpful.

Patrick

2013/5/17 Andrea Goijman <agoijman at cnia.inta.gov.ar>: