Hi All, I am trying to develop a GLM model in which the dependent data is a proportion data with 4 categories. I want to develop this model as an alternative method to Multinomial Logit Model for comparison of results. Here is the sample data:
Category SharesY x1 x2 1 0.797052154 7480 628 2 0.03590325 5178 1011 3 0.163265306 6711 1391 4 0.003779289 4177 1845 1 0.792581747 7562 628 2 0.051732552 5260 1011 3 0.150805271 6779 1391 4 0.004880429 5048 1693 . . . . . . . . . . . . By running this as a GLM model with family as binomial, and link function as logit I am getting a warning message: "In eval(expr, envir, enclos) : non-integer #successes in a binomial glm!". I am very new to R, and not sure whether it is even possible to run GLM model using proportion data but I read it somewhere and hence tried it. Any ideas/ thoughts/ alternate ways to run GLM using proportion data then please let me know. Thanks Prachi ______________________________________________ R-help@r-project.org mailing list 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.
