On Jun 23, 2006, at 5:22 AM, stefano iacus wrote:
Karl, this is something almost "impossible" (please not the quotes) to happen. The R GUI and the cmd line R version are the same R. Just an easy check: could you run sessionInfo() on both R with and without GUI? stefano
I would have thought so, and so I find this rather disturbing. In both cases I get the following: > sessionInfo() Version 2.3.1 (2006-06-01) i386-apple-darwin8.6.1 attached base packages: [1] "methods" "stats" "graphics" "grDevices" "utils" "datasets" "base" other attached packages: broman qtl "0.13-1" "1.03-3" If I detach all of the packages except "stats" and "base", I still have the same problem.
On Jun 23, 2006, at 4:43 AM, Joerg van den Hoff wrote:
by-the-by: the model is no good for the data: the exponential term is completely unidentifiable in the data which essentially follow a straight line (the result for `d' of 400% std. dev. shows this as well).
Right; I included just a bit of data that showed the problem. Regards, karl