binomial aggregate generalized mixed models
LOVE JASP- many thanks mto 1st rate team
BUT want to do generalized mixed models with binomial response, logit link
Works fine with 0,1 response variable in extended for (5210) observations, but not with same dara in more convenient aggreagate form (354) observations
Chosing binomila aggregate for input gvies following error message
"Generalized linear mixed model with binomial family and logit link function requres that the dependent variable is higher than 0 and lower than 1"
my data includes propotrions of 0 and 1.
aggregate runs perfectly in R
w3b2 <- glmer(cbind(freq, Nmax-freq) ~ W1_2_3*B2 +(W1_2_3|pno),
data=wb, family=binomial(link="logit"))
would be enormously grateful if jasp could be modified to accept start points p =0 or p=1
JASP is so marvellously easy to use, but with this one flaw
can provide inpur files if required
please help
Diana
Comments
This issue is continued on GitHub.
E.J.