Hi, I want to do 2-dimentional loess smoothing in gam through the command gam(Y~lo(X1,X2,span=span,degree=1) ) in library "gam". The span is the percentage of data points to define the neighborhood and used for the smoothing. Does this command do one 2-dimentional smoothing with neighboring defined according to the radius of the data point || X=(X1,X2) ||, or it does two 1-dimentional smoothing which smooths X1, X2 separately? I'm actually doing spatial smoothing with longitude and latitude, so I need the 2-dimentional smoothing so the results won't depend on the coordinate values. Thanks a lot!!! Best, Yijie Zhou
[[alternative HTML version deleted]] ______________________________________________ R-help@stat.math.ethz.ch 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.