search for: tt4663647

...t; 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. « [hide part of quote <http://r.789695.n4.nabble.com/predict-lm-tt4663647.html#>] I guess that, given the fact that you know how good/bad your models are, you can specify an additional error therm (a random variable with given mean and standard deviation). That variable will "cause" the predicted values to be different every time you predict the outcom...