Dear colleagues, I am struggling mightily with the mlogit package. First, the reason that I am using mlogit as opposed to multinom() in nnet is because my data is ranked, not just ordinal. So, I’m really trying to fit an exploded logit or rank-ordered model. All of the covariates of interest are individual-specific, none are alternative specific. The code below produces a model with my covariates of interest, so that is good. But, I cannot get predict.mlogit or effects.mlogit to work *at all*. The help package is quite unclear as to how to format the sample data that is fed to either of those two functions. Can any one help in that regard? Failing that, can anyone provide a suggestion for an alternative way of modelling ranked categorical data? I’m aware of the pmr and Rankcluster packages. The former however is also poorly documented and the latter is computationally intense to select clusters. I’m trying to do this as simply as possible while remaining loyal to the ranked structure of the data.
Thanks, Simon Kiss #Loadpackages library(RCurl) library(mlogit) library(tidyr) library(dplyr) #URL where data is stored dat.url<- 'https://raw.githubusercontent.com/sjkiss/Survey/master/mlogit.out.csv' #Get data dat<-read.csv(dat.url) #Complete cases only as it seems mlogit cannot handle missing values or tied data which in this case you might get because of median imputation dat<-dat[complete.cases(dat),] #Tidy data to get it into long format dat.out<-dat %>% gather(Open, Rank, -c(1,9:12)) %>% arrange(X, Open, Rank) #Create mlogit object mlogit.out<-mlogit.data(dat.out, shape='long',alt.var='Open',choice='Rank', ranked=TRUE,chid.var='X') #Fit Model mod1<-mlogit(Rank~1|gender+age+economic+Job,data=mlogit.out) ______________________________________________ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
