Hi Judit, Negative predictions are surprising indeed if your observed data are all positive values, and if you have a single trait. Do you have multiple traits? If they are correlated, information on one trait can push imputation of another in the negative range.
In any case, one option would be to log-transform the trait that has to take positive values. If it makes the trait more normally distributed, it’s generally a good thing to do. Then you can impute the missing values for the log-transformed trait, and finally transform back (take the exponential) if you need to. Cécile > On Jun 16, 2017, at 3:39 PM, Judit Mokos <mok...@gmail.com> wrote: > > Dear R-sig-phlyo members, > > I am using phylopars() in Rphylopars package in R to generate the missing > values in a large dataset about animal body traits (eg body size). ( > https://www.rdocumentation.org/packages/Rphylopars/versions/0.2.9/topics/phylopars) > This method is called imputation and what it does is to phylogenetically > estimate this missing datas. > > However the output of the imputation contains some negative values which > make no sense because all the estimated trait have to be bigger than zero. > I wonder how I can fix this issue or how to set up a minimum limit for the > estimated values. > > I'm not new in R but new in Rphylopars so maybe that question is pretty > naive but I couldn't find the solution. > > Thanks for your answer in advanced, > > Judit > > [[alternative HTML version deleted]] > > _______________________________________________ > R-sig-phylo mailing list - Rfirstname.lastname@example.org > https://stat.ethz.ch/mailman/listinfo/r-sig-phylo > Searchable archive at http://email@example.com/ _______________________________________________ R-sig-phylo mailing list - Rfirstname.lastname@example.org https://stat.ethz.ch/mailman/listinfo/r-sig-phylo Searchable archive at http://email@example.com/