Hello, I am modelling positive continuous data (including zeros) using the ZAGA distribution in GAMLSS and want to use the model for predictions. My final model includes smoothers (pb()) for the mu and nu parameter. First, I "blindly" used the default options for predictions but noticed that I do not have any zero values (or close to). Knowing this cannot be true, I learned that I also need the predictions for the other parameters (and not only mu as done by default), which I can extract e.g. with predictAll. My question is, how to "combine" all parameter values to calculate the expected value for one observation. It seems the function 'meanZAGA' does what I want, however not for new data. I tried to calculate the values I received with meanZAGA by hand in order to repeat it for predictions with new data but do not understand how to do it. I would appreciate any advise.
Thank you very very much! Cheers, Astrid -- Jetzt informieren: http://mobile.1und1.de/?ac=OM.PW.PW003K20328T7073a ______________________________________________ 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.
