Ok, I think I had a good idea to solve my problem and need someone who second me on that or tell me what a fool I am :-) .
My problem: I went for curve-fitting with nls() and took knot()-ecdf() for collecting data for nls-basis-dataframe. I came into trouble, because my x-y-data of the data.frame() was not "equal". I computed 1000 values and there were always some ties, those ties were not returned by knot()-function. Okay, I thought about it and now I am using 100 percentile-values for the data-frame - nls() works fine - moreover 100 percentiles are always there, exceptions won't happen. Is it wise to use 100 cumulative percentile for curve-fitting-base-data instead of ecdf()-function data? Thanks in advance. -- View this message in context: http://r.789695.n4.nabble.com/ecdf-to-nls-how-to-transform-data-tp3671754p3676932.html Sent from the R help mailing list archive at Nabble.com. ______________________________________________ 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.