I agree with Thomas that seeing a warning is a worry. If Paul's automap takes care of singular fits by discarding them, it should try to hide the warning for the user, because otherwise it will worry them. I see that the warning is issued by the C code in gstat, and will rethink about how a wrapper function at the R level should be able to suppress it.
Thanks for sharing automap with this list! -- Edzer Paul Hiemstra wrote: > Hi Thomas, > > The automatic fitting procedure checks a range of possible variogram > models, including Spherical, Exponential, Gaussian and Matern models. > Each of these models could lead to a singular model fit. The fitting > function in gstat (fit.variogram) says the following on singular model fits: > ... > _______________________________________________ R-sig-Geo mailing list R-sig-Geo@stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-geo