Robert van Geldern <[EMAIL PROTECTED]> wrote: > I used the Locift package for smoothing xy-data using R [...]
> Does anybody how to geht the xy-Values of the confidence band out of the > program to a table? > Getting the data of the fit at any point x is no problem. It seems that > the function calculates the standard error at each ponit, but is there a > way to get the values of the conf. intervals at a point x directly? I think the predict.locfit function is what you need. Do ``?predict.locfit'' at the command prompt -- I think you just need to specify ``se.fit=T'' to get the values for the error bars. Hope this helps, Robert Dodier -- I rarely saw people sitting at computers producing real code wearing ties. -- Philippe Kahn . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
