Perhaps something like this: X = jitter(1:10) Y = jitter(3*X-5, factor = 3) X[3] = NA m = lm(Y~X)$fitted.values
fits <- rep(NA, length(X)); fits[as.numeric(names(m))] <- m; cbind(X,Y,fits) Michael On Tue, Nov 1, 2011 at 8:52 AM, Johannes Radinger <jradin...@gmx.at> wrote: > Hi, > > I have got a dataset with > the variables Y,X1,X2,X3. > Some of these variables contain NAs. Therefore > incomplete datasets aren't recognized when > I am doing a regression like: > > model <- lm(Y~X1+X2+X3) > > so the resulting vector of resid(model) is > obviousely shorter then the original variables. > > How can I combine the residuals-vector with > the original dataset (Y,Xi,...). I recognize > that the residuals give also a kind of an index > like from the row names (2,5,7,8,9,12)... > > Is it possible to use in such a case the > data.frame command? Or what is the best > way to attach the resulting residuals back to > the original dataframe with incomplete datasets? > > thanks > > /Johannes > -- > > ______________________________________________ > 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. > ______________________________________________ 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.
