search for: lqs_fitlots

...-1e-20) and so sqrt complains. Now I have no idea what ltsreg is doing, and what z$crit is. The letters suggest a z critical value, which can't be negative. Certainly, the programmer who decided to stuff it into sqrt() didn't think it would be negative. So this suggests a problem in the lqs_fitlots function call, but I haven't looked at that. I understand that the problem may be that bootstrapping such a small data set is insane, but I'm just copying Efron and Tibshirani (who do this with lmsreg). So the problem may be only that more informative error messages are needed (or failure...

Since this has come up already: It is a good idea to re-install all packages for a minor-version increment of R, e.g. 2.5.1 -> 2.6.0 (it is major.minor.patchlevel). This is most easily done by > update.packages(checkBuilt=TRUE, ask=FALSE) However, if you don't want to do that yet, be aware that - Certain S4-using packages must be reinstalled, and using old versions can make R

Le Sat, Nov 18, 2017 at 03:26:53PM -0800, Bill Harris a ?crit : > Incidentally, you can see a bit more complete description at > https://unix.stackexchange.com/questions/402560/how-do-i-install-r-on-debian-stretch-given-the-r-api-3-issue > . Hi Bill and everybody, if one installs R >= 3.4.2 from any source, then some Debian packages will be broken. The change of r-api virtual